Energy control system and method for ship application

Abstract

The invention discloses an energy control system and method for ship application. The system comprises a high-voltage main loop control module, a data acquisition module, an intelligent control module, an expansion and protection module, a multi-stage power electronic conversion module and a heat dissipation module. The high-voltage main loop control module comprises a high-voltage contactor, a fuse and a pre-charging circuit; the data acquisition module comprises a current sensor, a temperature sensor and a battery management system interface; the intelligent control module comprises a BMS module control and communication unit and is used for executing an energy optimization algorithm with the purposes of minimizing energy consumption and maximizing efficiency and interacting with an upper energy management system. The expansion and protection module comprises a high-voltage connector, a bus bar and a metal box body and is used for accessing the multi-energy module; the multi-stage power electronic conversion module is electrically connected to the high-voltage direct-current bus, and adjusts the duty ratio and the output power based on an energy optimization result; and the heat dissipation module is linked with the temperature sensor to dissipate heat of key components so as to maintain the operation temperature.

Description

Technical Field

[0001] The present invention relates to the technical field of ship energy management, and in particular to an energy control system and method for ship applications. Background Art

[0002] As the global shipping industry transitions toward low-carbon and intelligent shipping, ship energy systems are upgrading from traditional single-diesel engine drives to a comprehensive energy supply model that complements lithium batteries, supercapacitors, and diesel generators to meet the needs of reducing carbon emissions, improving energy efficiency, and adapting to complex operating conditions (such as starting, cruising, and docking). However, existing ship energy control systems have exposed the following technical bottlenecks in practical applications: 1. Inefficient multi-energy collaborative control and weak dynamic adaptability Traditional systems often employ fixed power allocation strategies, making it difficult to respond in real time to the highly nonlinear and sudden changes in a ship's propulsion load (affected by ship speed, propeller speed, and draft). Furthermore, key parameters such as the state of health (SOH) and state of charge (SOC) of energy modules (such as lithium batteries and supercapacitors) lack dynamic monitoring and optimized scheduling mechanisms. This results in low energy conversion efficiency (less than 80% under certain operating conditions) and redundant energy consumption (backup energy modules are often operating inefficiently).

[0003] 2. Insufficient startup surge suppression In existing energy control systems, when the ship's high-voltage DC bus is started, the instantaneous charging of the bus capacitor will generate a surge current impact.

[0004] 3. Limited data collection and processing accuracy affects control decisions (1) Severe interference in current data: Strong electromagnetic interference during the operation of the ship's high-voltage system (such as propellers and inverters) causes high-frequency noise in the raw data of the current sensor. Traditional fixed window filtering methods are difficult to balance steady-state smoothness and dynamic responsiveness, which directly affects the calculation accuracy of the energy optimization algorithm. (2) Large temperature monitoring error: The nonlinear characteristics of sensors such as NTC thermistors and K-type thermocouples introduce original errors, and the ambient temperature fluctuations in the high-voltage control box cause the "component surface temperature" measurement value to contain environmental interference; (3) Inaccurate battery status assessment: Traditional state of charge (SOC) calculation relies on a single ampere-hour integration method or open circuit voltage (OCV) correction at a fixed temperature, and does not consider the impact of battery aging (SOH decline) on OCV characteristics, leading to misjudgment of energy scheduling.

[0005] 4. Energy optimization algorithm has poor adaptability Traditional PID control relies solely on the current error to adjust the output, lacks prediction of future load changes, and is difficult to dynamically adjust the priority of energy modules according to operating conditions, resulting in low energy utilization efficiency.

[0006] In summary, the existing ship energy control system has significant defects in multi-energy collaborative optimization, startup reliability, safety protection redundancy, data acquisition accuracy and energy scheduling adaptability. There is an urgent need for a new energy control system and method that can achieve efficient energy scheduling, highly reliable pre-charging control, multi-level active protection and precise processing of multi-source data. Summary of the Invention

[0007] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide an energy control system and method for ship applications, which is used to solve the technical problems existing in the existing ship energy control system in terms of multi-energy synergy efficiency, startup reliability, safety protection redundancy, data acquisition accuracy and energy scheduling adaptability.

[0008] In order to solve the above problems, the technical solutions adopted by the present invention are as follows: An energy control system for ship applications, comprising: The high-voltage main circuit control module includes a high-voltage contactor, a fuse, and a pre-charging circuit. The high-voltage contactor is used for on-off control of the main circuit, the fuse is used for overcurrent protection, and the pre-charging circuit includes a current-limiting resistor, an auxiliary contactor, and a control logic circuit for pre-charging when the system starts. Data acquisition module, including current sensor, temperature sensor and battery management system interface, used to collect main circuit current, key component temperature and battery status data respectively; an intelligent control module, including a BMS module control and communication unit, configured to execute an energy optimization algorithm with minimizing system energy consumption and maximizing energy conversion efficiency as objective functions, and to interact with a higher-level energy management system; Expansion and protection module, including high-voltage connectors, busbars and metal boxes, for accessing multiple energy modules; a multi-stage power electronic conversion module electrically connected to the high-voltage DC bus and configured to adjust the duty cycle and output power based on the energy optimization result; The heat dissipation module is linked with the temperature sensor to dissipate heat from key components to maintain the system operating temperature.

[0009] An energy control method for ship applications comprises the following steps: Start the system, connect multiple energy modules through the expansion and protection module and communication protocol, and pre-charge the bus capacitor using the current-limiting resistor of the pre-charge circuit; The main circuit current data is monitored in real time through the current sensor, the temperature data of key components is collected through the temperature sensor, and the battery cell voltage, temperature, state of charge and health status data are obtained through the battery management system interface; Based on the collected data, the BMS module control and communication unit executes a dynamic energy optimization algorithm. Combined with the multi-stage power electronic conversion module, it dispatches energy for the 1000V high-voltage DC bus, coordinates charging and discharging strategies, and controls voltage fluctuations. When overcurrent, overvoltage or overtemperature anomalies are detected, the main circuit is quickly cut off through the high-voltage contactor, and the fuse is triggered to blow and isolate the fault according to the cut-off result.

[0010] Preferably, the real-time monitoring of current data includes: The current sensor is installed in series on the busbar of the high-voltage main circuit to collect current data in real time at a preset sampling frequency; The collected current data is transmitted to the intelligent control module in real time and processed using an adaptive sliding average filtering strategy that includes dynamic evaluation of noise levels, adaptive adjustment of filtering parameters, optimization of operating condition associations, and isolation of abnormal data. It is then used for dynamic calculation of the energy optimization algorithm and real-time determination of overcurrent faults.

[0011] Preferably, when collecting temperature data, the following steps are included: The temperature sensors are respectively integrated on the surface of the high-voltage contactor contact, the fuse core and the current-limiting resistor of the pre-charge circuit, and the temperature data is collected in real time at a preset sampling frequency; The collected temperature data is transmitted to the intelligent control module in real time. After being processed by segmented linearization correction and ambient temperature compensation, it is used to trigger the start-up of the heat dissipation module and serve as the basis for determining over-temperature faults. Among them, the piecewise linearization correction and ambient temperature compensation processing include: piecewise linearization correction of nonlinear characteristics, dynamic compensation of ambient temperature field and self-correction of thermal resistance aging.

[0012] Preferably, obtaining battery cell data includes: The battery cell voltage is measured through the BMS's built-in 16-bit high-precision analog-to-digital converter; The battery cell temperature is collected through the NTC thermistor sensor integrated in the BMS; The state of charge is calculated by the ampere-hour integration method combined with open circuit voltage correction; Comprehensively evaluate the health status through battery internal resistance increment analysis and capacity decay model; The acquired battery cell voltage, temperature, state of charge, and health status data are transmitted to the intelligent control module in real time through the battery management system interface. After CRC verification and outlier filtering, they are used in the energy optimization algorithm. Among them, the calculation of the state of charge includes: dynamic stable operating condition identification and correction triggering, temperature compensated OCV-SOC mapping table construction, aging-adaptive OCV correction, OCV denoising processing under vibration interference, multi-stage correction weight allocation and SOC error closed-loop correction.

[0013] Preferably, a comprehensive assessment of health status includes: Select characteristic frequency points to inject sinusoidal current signals, synchronously collect the battery terminal voltage response, and calculate the AC impedance at each frequency point through fast Fourier transform; Calculate the incremental rate of AC impedance at each frequency point relative to the initial value, and construct an internal resistance aging correlation model based on the incremental rate and weight coefficient; The temperature acceleration factor is calculated based on the Arrhenius equation, and the equivalent full charge and discharge cycle number is obtained using the acceleration factor and the actual cycle number; A capacity decay model is constructed based on the number of equivalent full charge and discharge cycles, initial capacity, and decay constant. The weight coefficients of the internal resistance aging association model and the capacity attenuation model are dynamically adjusted according to the current SOH value, and the final health status is output.

[0014] Preferably, the dynamic energy optimization algorithm adopts a model predictive control framework, calculates the optimal energy distribution ratio of each energy module within the optimization cycle by inputting real-time main circuit current, key component temperature, battery cell state of charge, health status and bus voltage data, and generates corresponding MPC instructions; The isolated bidirectional DC-DC converter performs bidirectional energy conversion between the energy module and the bus, and adjusts the duty cycle according to the MPC instructions; The grid-connected DC-AC inverter converts the bus DC energy into AC load-usable electrical energy and adjusts the output power according to the MPC instructions; Among them, the model predictive control framework includes: a dynamic load prediction model that dynamically adjusts the prediction time domain length according to the load change rate, a multi-objective weight dynamic allocation strategy, life constraint and temperature safety boundary embedding, multi-energy module collaborative optimization, and online model self-update.

[0015] Preferably, when an overcurrent, overvoltage or overtemperature anomaly is detected, the high-voltage contactor is triggered to disconnect first, and the fuse blow countdown is started at the same time; If the high-voltage contactor fails to disconnect, the fuse will melt within the set time to achieve backup isolation. If the high-voltage contactor successfully disconnects, the fuse will remain inactive.

[0016] Preferably, when the system starts, the connected energy modules are scanned through broadcast frames to receive device descriptors fed back by the energy modules; The intelligent control module detects the voltage range, communication protocol compatibility and protection level of the energy module; After the matching is successful, a unique device ID is assigned to the energy module, and the physical connection with the high-voltage DC bus is completed by closing the auxiliary contactor.

[0017] Preferably, when the system is started, the auxiliary contactor is closed and the current limiting resistor is connected in series to the main circuit; The bus capacitor begins to charge through the current-limiting resistor, and the control logic circuit monitors the voltage across the capacitor in real time; When the capacitor voltage or pre-charge time reaches the set value, the control logic circuit determines that the pre-charge is completed, closes the main circuit high-voltage contactor, and opens the auxiliary contactor at the same time to complete the main circuit switching.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. The efficiency of multi-energy coordinated control is significantly improved, reducing system energy consumption and fluctuations A dynamic energy optimization algorithm within the Model Predictive Control (MPC) framework combines real-time data such as ship speed and propeller speed to predict future load demands. The weighting of the objectives of "minimizing energy consumption" and "maximizing efficiency" is dynamically adjusted based on the starting, cruising, and docking conditions. Furthermore, a priority matrix based on operating conditions is designed based on the differences in the characteristics of lithium batteries, supercapacitors, and diesel generators to achieve dynamic power allocation among multiple energy modules.

[0019] 2. The startup surge suppression capability is enhanced, and the system reliability is greatly improved The design includes a pre-charge circuit with a current-limiting resistor, auxiliary contactor, and control logic circuit. It monitors the bus capacitor voltage rise rate and pre-charge time in real time and completes the main circuit switching when the capacitor voltage or pre-charge time reaches the preset value. This design effectively suppresses the startup inrush current.

[0020] 3. Improved multi-level active safety protection mechanisms and enhanced fault isolation redundancy The coordinated control strategy of "high-voltage contactor priority disconnection + fuse backup isolation" is adopted: when the high-voltage contactor detects overcurrent, overvoltage or overtemperature anomalies, it cuts off the main circuit; if the contactor fails to disconnect (such as contact adhesion), the fast fuse will melt within 5ms to provide backup protection.

[0021] 4. The accuracy of multi-source data collection and processing has been greatly improved, supporting precise control decisions. Current data: Adopting an adaptive sliding average filtering strategy, the noise suppression effect under steady-state conditions is improved, the response delay under dynamic conditions is shortened, abnormal spikes are effectively eliminated, and reliable input is provided for the energy optimization algorithm.

[0022] Temperature data: Through piecewise linearization correction, dynamic compensation for ambient temperature, and thermal resistance aging self-correction, the temperature monitoring accuracy of key components is improved, providing a reliable basis for heat dissipation control and over-temperature fault judgment.

[0023] Battery status: The multi-method fusion of SOC calculation (ampere-hour integration + temperature-compensated OCV correction + aging adaptive correction) significantly reduces the error in state-of-charge calculation; the multi-stage weighted fusion of SOH assessment (internal resistance increment method + capacity decay model) obtains a more accurate battery health status, providing precise data support for energy scheduling and life prediction.

[0024] 5. Optimize energy scheduling adaptability and multi-objective balance capabilities to meet complex working conditions The dynamic load prediction model under the MPC framework (combined with data such as ship speed and propeller speed) can predict load changes in advance, allowing control instructions to proactively adapt to operating conditions. By dynamically allocating weights to operating conditions, it resolves the multi-objective contradictions of traditional control. At the same time, it embeds battery life constraints, temperature safety boundaries, and a multi-energy priority matrix to achieve the "foresight-safety-efficiency" synergy of energy scheduling.

[0025] In summary, the present invention significantly improves the reliability, efficiency and safety of ship energy systems through the deep integration of system-level hardware design and intelligent algorithms, providing key technical support for the green and intelligent transformation of ships.

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 1 is a precharge circuit diagram of an embodiment of the present invention; Figure 2 This is a front view of a high-voltage control box according to an embodiment of the present invention; Figure 3 1. is a top view of a high-voltage control box according to an embodiment of the present invention; Figure 4 is a side view of a high-voltage control box according to an embodiment of the present invention; Figure 5 This is a step diagram of an energy control method for ship applications according to an embodiment of the present invention.

[0028] Explanation of the accompanying figures: 1. Current limiting resistor; 2. Auxiliary contactor; 3. Control logic circuit; 4. High-voltage contactor; 5. Fuse. DETAILED DESCRIPTION

[0029] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0030] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0031] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0032] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0033] Example 1 provides an energy control system for ship applications, including: a high-voltage main circuit control module, a data acquisition module, an intelligent control module, an expansion and protection module, a multi-stage power electronic conversion module and a heat dissipation module.

[0034] See Figure 1 The circuit diagram of the high-voltage main circuit control module shows that the high-voltage main circuit control module includes a high-voltage contactor 4, a fuse 5 and a pre-charging circuit, wherein the high-voltage contactor 4 is used for main circuit on-off control, the fuse 5 is used for overcurrent protection melting, and the pre-charging circuit includes a current limiting resistor 1, an auxiliary contactor 2 and a control logic circuit 3, which is used for pre-charging when the system starts.

[0035] The data acquisition module includes a current sensor, a temperature sensor and a battery management system (BMS) interface, which are used to collect main circuit current, key component temperature and battery status data respectively.

[0036] The intelligent control module includes a BMS module control and communication unit, supports CAN, RS485, and Ethernet protocols, and is configured to execute energy optimization algorithms, safety protection strategies, and data communications with the objective function of minimizing system energy consumption and maximizing energy conversion efficiency, and interact with the upper-level energy management system (EMS).

[0037] The expansion and protection module includes high-voltage connectors, busbars, and a metal box (dust-proof, waterproof, corrosion-resistant, and fire-proof). It uses shielded cables and a grounding design to enable multi-energy module access, electromagnetic interference suppression, and physical protection.

[0038] The multi-stage power electronic conversion module is electrically connected to the high-voltage DC bus and is configured to adjust the duty cycle and output power based on the energy optimization result to achieve efficient energy conversion.

[0039] The heat dissipation module works in conjunction with the temperature sensor to dissipate heat from key components to maintain the system operating temperature.

[0040] See Figures 2 to 4 In a possible embodiment, the energy control system of this embodiment is integrated into a high-voltage control box.

[0041] In one possible embodiment, the current sensor uses a Hall sensor or a shunt. The Hall sensor supports a wide range of measurement from 0 to 500A, with a measurement accuracy of ±0.5% FS (full-scale error) and a response time of ≤10ms. The shunt is made of low-impedance manganese-copper alloy and is combined with a high-precision differential amplifier to realize current signal conversion.

[0042] In a possible embodiment, the temperature sensor uses an NTC thermistor or a K-type thermocouple. The measurement range of the NTC thermistor is -40°C to 125°C, the accuracy is ±1°C, and the response time is ≤20ms. The measurement range of the K-type thermocouple is -200°C to 800°C. The heat dissipation module is a fan or a liquid cooling system.

[0043] In a possible embodiment, the high-voltage contactor 4 is a high-voltage contactor 4 that supports a 1000V DC voltage and a main contact breaking time ≤10ms, and the fuse 5 is a fast fuse with a breaking capacity ≥20kA, whose time-current characteristics match the breaking time of the high-voltage contactor 4.

[0044] In one possible embodiment, the multi-stage power electronic conversion module includes: an isolated bidirectional DC-DC converter and a grid-connected DC-AC inverter; the isolated bidirectional DC-DC converter is used for bidirectional energy conversion between the energy module and the bus, and adjusts the duty cycle according to MPC instructions; the grid-connected DC-AC inverter is used to convert the bus DC energy into usable electrical energy for the AC load, and adjusts the output power according to MPC instructions.

[0045] In a possible embodiment, the high-voltage connector is a rectangular plug-in high-voltage connector (compliant with IEC 60529 standard), the interface has an IP67 protection grade, can withstand a 1000V DC voltage, and the contact surface is silver-plated (contact resistance ≤ 50μΩ); the high-voltage connector is equipped with a mechanical locking structure (to prevent loosening due to ship vibration), supports blind insertion positioning (positioning pin deviation ≤ 0.5mm), and is suitable for the physical connection of multiple types of energy modules such as lithium battery packs, supercapacitors, and diesel generator inverters.

[0046] Example 2, see Figure 5 The energy control method for ship applications is shown in the following steps. Figure 5 An energy control method for ship applications is shown, comprising the following steps: Step S1: Module access and pre-charging: Start the system and access multiple energy modules through expansion and protection modules and communication protocols (CAN, RS485, Ethernet) to achieve compatibility expansion for large-scale deployment; The bus capacitor is precharged through the current limiting resistor 1 of the pre-charging circuit to suppress the inrush current impact when the system starts; Step S2: Data acquisition: The current data of the ship's high-voltage main circuit is monitored in real time through the current sensor, the temperature data of key components in the high-voltage control box is collected through the temperature sensor, and the battery cell voltage, temperature, state of charge (SOC) and state of health (SOH) data are obtained through the battery management system (BMS) interface; Step S3: Algorithm optimization and energy scheduling: Based on the collected data, the dynamic energy optimization algorithm is executed through the BMS module control and communication unit. Combined with the multi-stage power electronic conversion module, the 1000V high-voltage DC bus is energy-scheduled to complete the coordination of charging and discharging strategies, energy distribution of different energy modules, and voltage fluctuation control (fluctuation range ≤±5%). Step S4, active safety protection: when overcurrent, overvoltage or overtemperature anomalies are detected, the main circuit is quickly cut off through the high-voltage contactor 4, and the fuse 5 is triggered to melt and isolate the fault according to the cut-off result.

[0047] In the above step S2, the real-time monitoring of the ship's high-voltage main circuit current data includes: The current sensor is installed in series on the busbar of the high-voltage main circuit to collect current data in real time at a sampling frequency of 100 Hz; The collected current data is transmitted to the intelligent control module in real time via the CAN bus with a period of ≤20ms. After being processed by an adaptive sliding average filtering strategy, it is used for dynamic calculation of the energy optimization algorithm and real-time determination of overcurrent faults.

[0048] Background: In ship energy control systems, the real-time and accuracy of high-voltage main circuit current data is crucial for energy optimization algorithms and fault diagnosis. However, the ship's operating environment is complex, and current data is susceptible to the following interference and challenges: Strong noise interference: Ship high-voltage systems (such as thrusters and inverters) generate strong electromagnetic interference during operation, causing high-frequency noise (such as surges and spikes) in the raw data collected by the current sensor, which directly affects the dynamic calculation accuracy of the energy optimization algorithm.

[0049] Dynamic changes in operating conditions: Ship operation includes multiple operating conditions such as startup (rapid current increase), cruising (stable current), and berthing (current decrease). Different operating conditions have conflicting requirements for the "smoothness" and "real-time" of current data: stable operating conditions require noise suppression (smoothness priority), while load mutation conditions require rapid response (real-time priority). The fixed window filtering method cannot take both into account.

[0050] Abnormal data pollution: Ship vibration or sensor transient failure may cause abnormal spikes in current data (such as instantaneous values ​​exceeding 200% of the rated current). If not effectively eliminated, it will pollute the filtering results and falsely trigger overcurrent protection. Based on this: In a possible embodiment, the adaptive sliding average filtering strategy processing includes: Dynamic evaluation of noise level: Real-time calculation of the variance σ of the current data within the current sampling window 2 and the rate of change of adjacent sampling points ΔI / Δt, where the variance σ 2 Reflects the degree of data fluctuation, and the rate of change ΔI / Δt reflects the current mutation characteristics; Adaptive adjustment of filter parameters: when the variance σ 2 > Set the threshold (such as 5A 2 ) or the rate of change |ΔI / Δt| > a set threshold (e.g., 20A / ms) (corresponding to sudden changes in ship load or strong electromagnetic interference), increase the sliding average window size (e.g., from the default 5 points to 10 points) and increase the weight coefficient of historical data (e.g., set the weight of the most recent 3 points to 0.4, the middle 3 points to 0.3, and the earliest 4 points to 0.3) to enhance the suppression of high-frequency noise; When the variance σ 2 When the current is less than or equal to the set threshold and the rate of change |ΔI / Δt| is less than or equal to the set threshold (corresponding to a stable ship cruising scenario), reduce the sliding average window size (for example, from 5 points to 3 points) and reduce the weight of historical data (for example, the weight of the most recent two points is 0.6, and the weight of the earliest point is 0.4) to improve the response speed of the filtered data to current changes; Working condition correlation optimization: The intelligent control module obtains the ship's operating status (starting / cruising / docking). During the starting condition (current rising phase), an additional feedforward compensation term is introduced (the compensation value is the current change rate of the previous cycle multiplied by the current sampling interval) to avoid current peak lag caused by window expansion. During the docking condition (current falling phase), the compensation coefficient is reduced to prevent numerical deviation caused by overcompensation. Abnormal data isolation: If the current value of a single sampling point exceeds 200% of the rated current and the rate of change is greater than 50A / ms (determined as spike interference), the point is marked as an outlier and removed. Linear interpolation of the two points before and after is used to replace it to prevent the outlier from contaminating the filtering results.

[0051] It is necessary to further explain in the embodiments of the present invention that the above processing process achieves the following by dynamically adjusting the filtering parameters (window size, data weight) and combining the working condition identification and anomaly isolation: Steady-state conditions (such as cruising): suppress high-frequency noise and reduce steady-state errors; Dynamic working conditions (such as startup): reduce response delay and avoid peak lag; Abnormal scenarios (such as vibration interference): eliminate abnormal points to ensure data reliability.

[0052] The adaptive sliding average filtering strategy provided in this embodiment is aimed at current data collection under complex ship operating environments, and achieves a balance between "smoothness and real-time performance" through dynamic parameter optimization.

[0053] In the above step S2, when collecting the temperature data of key components in the high-voltage control box, it includes: The temperature sensors are integrated on the contacts of the high-voltage contactor 4, the fuse core 5 and the surface of the current-limiting resistor 1 of the pre-charge circuit, and the temperature data is collected in real time at a sampling frequency of 50 Hz; The collected temperature data is transmitted to the intelligent control module in real time via the CAN bus with a cycle of ≤30ms. After piecewise linearization correction and ambient temperature compensation, it is used to trigger the start of the heat dissipation module (for example, when the temperature is ≥75°C, heat dissipation is started) and serves as the basis for determining over-temperature faults (for example, the over-temperature threshold is set to 100°C). Among them, the temperature sensor adopts NTC thermistor or thermocouple.

[0054] Background: In ship energy control systems, temperature monitoring of key components within the high-voltage control box (such as the high-voltage contactor 4 contacts, fuse 5 fuse core, and current-limiting resistor 1) is crucial for ensuring safe system operation (e.g., triggering the cooling module and determining over-temperature faults). However, temperature data collection faces the following challenges: The NTC thermistors and K-type thermocouples commonly used in ships have significant nonlinear characteristics: The resistance-temperature (RT) curve of NTC decays exponentially. Directly inferring the temperature from the resistance value will introduce an original error of ±1°C. The thermoelectric potential-temperature (ET) curve of the K-type thermocouple is nonlinear over a wide temperature range (the Seebeck coefficient is not constant), and the original measurement error can reach ±2°C.

[0055] Ambient temperature field interference: The ambient temperature inside a ship's high-voltage control box is easily affected by equipment operation (such as busbar heating) and external climate (such as cabin temperature fluctuations), causing the "component surface temperature" collected by the sensor to include ambient temperature interference. For example, when the ambient temperature rises from 25°C to 40°C, the measured contact surface temperature may be overestimated by 1.5°C due to ambient heat conduction (assuming a thermal resistance of 0.1°C / W), failing to accurately reflect the component's actual temperature rise (the actual temperature may only increase by 10°C, but the measured value will show an increase of 11.5°C).

[0056] Thermal resistance drift due to component aging: After long-term operation, oxidation of the high-voltage contactor 4 contacts and aging of the fuse 5 core can change the equivalent thermal resistance (R_th) between the component and the environment. For example, an initial contact thermal resistance of 0.1°C / W may increase to 0.2°C / W after 1000 hours of operation. Without dynamic correction of thermal resistance parameters, the ambient temperature compensation model will gradually become ineffective, and measurement errors will accumulate over time, threatening the long-term reliability of the system. Based on this: In a possible embodiment, the piecewise linearization correction and ambient temperature compensation process includes: Nonlinear characteristic piecewise linearization correction: For NTC thermistors, based on their resistance-temperature (RT) characteristic curve (B-value coefficient 3435K, resistance 10kΩ at 25°C), the measurement range (-40°C to 125°C) is divided into three intervals: low temperature (-40°C to 25°C), medium temperature (25°C to 80°C), and high temperature (80°C to 125°C). Within each interval, the RT data of three sets of standard temperature points (e.g., -40°C, 0°C, 25°C; 25°C, 50°C, 80°C; and 80°C, 100°C, 125°C) are fitted using the least squares method. Linearization equations are then established for each of these intervals. After correction, the measurement error is reduced from ±1°C to ±0.5°C. For K-type thermocouples, based on their thermoelectric potential-temperature (ET) characteristics (Seebeck coefficient approximately 41 μV / °C), the measurement range (-200°C to 800°C) is divided into four intervals (-200°C to 0°C, 0°C to 400°C, 400°C to 600°C, and 600°C to 800°C). A lookup table combined with linear interpolation (e.g., in the 0°C to 400°C range, five standard points are used: 0°C (0 mV), 100°C (4.096 mV), 200°C (8.138 mV), 300°C (12.209 mV), and 400°C (16.397 mV)) is used to convert the nonlinear thermoelectric potential signal into a linear temperature value. After correction, the error is optimized from ±2°C to ±1°C. Dynamic compensation of ambient temperature field: Arrange three auxiliary temperature sensors (using the same type of NTC or thermocouple as the measured point) in the non-heating area of ​​the high-voltage control box (such as the side wall of the metal box and the non-current-carrying section of the busbar) to collect the ambient temperature T in real time. env1 、T env2 、T env3 , calculate the average ambient temperature T envavg = (T env1 +T env2 +T env3 ) / 3; Based on the thermal conductivity characteristics of the key components under test (high-voltage contactor 4 contacts, fuse 5 fuse core, current limiting resistor 1), an ambient temperature impact model is established: T measured =Tactual +R_th×(T envavg -T ref ); Among them, T measured is the original measurement value of the sensor, T actual is the actual temperature of the component, R_th is the equivalent thermal resistance between the component and the environment (contact R_th1=0.1℃ / W, fuse R_th2=0.08℃ / W, resistor R_th3=0.15℃ / W), T ref is the initial ambient temperature of the system (25°C); By real-time calculation of T actual =T measured -R_th×(T envavg -T ref ) to eliminate the interference of ambient temperature fluctuations on the measured value (for example, when the ambient temperature rises from 25°C to 40°C, the contact measurement value can be compensated to decrease by 1.5°C (0.1°C / W×15°C)); Thermal resistance aging self-correction: The intelligent control module records the system's cumulative operating time t (unit: hour) and the historical temperature data of the measured components, and fits the equivalent thermal resistance R using the least squares method. th The coefficient k of change over time t (e.g. k = 0.001°C / (W·h)) dynamically updates the thermal resistance parameters: R_th(t) = R_th initial + k×t; Among them, R_th initial is the initial thermal resistance, R_th1 initial is the initial thermal resistance of the contact, R_th2 initial is the initial thermal resistance of the fuse, R_th3 initial is the initial thermal resistance of the resistor; (For example, the initial thermal resistance of the contact R_th1 initial =0.1℃ / W, updated to R_th1(1000)=0.1+0.001×1000=0.2℃ / W after 1000 hours of operation), to compensate for the increase in thermal resistance caused by contact oxidation and fuse core aging.

[0057] In the embodiments of the present invention, further explanation is required: the aforementioned processing utilizes nonlinear segmented correction to linearize the sensor characteristic curve piecewise, and reduces the original measurement error through least squares or table lookup methods; an auxiliary sensor is used to collect the ambient temperature within the chamber, establishing a heat conduction model to separate the component's internal temperature rise from environmental interference; operating time and historical temperature data are recorded, and thermal resistance parameters are dynamically updated to avoid compensation failures due to component aging. In summary, through error correction, interference isolation, and model self-updates, the steady-state accuracy and dynamic response capability of temperature data are improved, providing a reliable basis for heat dissipation control and overtemperature fault determination, thereby ensuring the safe and efficient operation of the energy control system.

[0058] In the above step S2, obtaining the battery cell voltage, temperature, state of charge and health status data includes: The battery cell voltage is measured using the BMS's built-in 16-bit high-precision analog-to-digital converter (ADC), with a measurement range of 2.5V to 4.2V (suitable for lithium-ion batteries), a measurement accuracy of ±10mV, and a single-cell sampling interval of ≤100ms. The battery cell temperature is collected through the NTC thermistor sensor integrated in the BMS. The NTC thermistor sensor is attached to the surface of the battery cell in a one-to-one correspondence with the battery cell. The measurement range is -20℃~60℃, the accuracy is ±2℃, and the sampling frequency is 5Hz. The state of charge (SOC) is calculated by the ampere-hour integration method combined with open circuit voltage (OCV) correction; Comprehensively evaluate the state of health (SOH) through battery internal resistance increment analysis (based on AC impedance method) and capacity fade model (combining cycle number and temperature stress data); The acquired battery cell voltage, temperature, state of charge (SOC) and state of health (SOH) data are transmitted to the intelligent control module in real time at a cycle of ≤50ms through the battery management system (BMS) interface using the CAN bus communication protocol (supporting a rate of 500kbps) or the Ethernet protocol (supporting a rate of 100Mbps). After CRC verification and outlier filtering, they are used to adjust the charge and discharge power of the energy optimization algorithm and trigger the active safety protection strategy.

[0059] Background: In ship energy control systems, state of charge (SOC) is a core parameter for battery energy management, directly affecting energy scheduling strategies (such as charge and discharge power allocation), safety protection (such as avoiding overcharging and overdischarging), and life prediction. However, the operating environment of ship batteries is complex, and traditional single SOC calculation methods (such as the ampere-hour integration method or the OCV method) have significant limitations: The cumulative error problem of the ampere-hour integration method: The ampere-hour integration method calculates SOC by integrating current, offering high real-time performance. However, errors accumulate over time (e.g., current sensor accuracy deviation and the temperature dependence of the coulombic efficiency η). During long-term battery operation (e.g., cruising for several hours), errors can exceed 5%, leading to misjudgments in energy scheduling (e.g., prematurely activating a diesel generator due to an incorrect assessment of insufficient remaining battery charge).

[0060] Effect of temperature on OCV-SOC relationship: Battery OCV varies significantly with temperature (for example, the OCV of a lithium-ion battery at -20°C is approximately 0.2V lower than at 25°C). If you directly use an OCV-SOC mapping table at a fixed temperature, the SOC correction error under low or high temperature conditions may exceed 10%, leading to overcharging risks (e.g., misjudging the SOC as high at low temperatures and continuing to discharge even when the actual capacity is insufficient).

[0061] Changes in OCV characteristics due to battery aging: As the battery's state of health (SOH) decreases, the slope of the OCV-SOC curve decreases and the intercept shifts (for example, when SOH is less than 80%, the OCV corresponding to the same SOC is 0.1V lower than that of a new battery). If the OCV mapping table is not dynamically corrected, the SOC calculation error of an aging battery will increase as the SOH decreases, affecting life management strategies (such as misjudging the battery's remaining life).

[0062] Voltage disturbance caused by ship vibration: Mechanical vibrations during ship navigation can cause instantaneous fluctuations in battery terminal voltage, directly affecting the accuracy of OCV measurements. Traditional filtering methods (such as fixed window averaging) may over-smooth the effective signal or retain interference noise, resulting in deviations in the OCV correction value. Based on this: In a possible embodiment, calculating the state of charge includes: Dynamic stable working condition identification and correction triggering: The intelligent control module monitors the battery current and operating status in real time and triggers the OCV correction process when the following conditions are met: The battery is in a static or low-current operating state (charge and discharge current ≤ 0.1C, where C is the rated capacity of the battery) and lasts for ≥ 2 minutes (eliminating the impact of load fluctuations on the terminal voltage); Or ampere-hour integral cumulative error warning (when the cumulative running time is ≥15 minutes and no correction is made, correction is triggered compulsorily); Temperature compensated OCV-SOC mapping table construction: The OCV-SOC benchmark curves at different temperatures (-20°C, 0°C, 25°C, 40°C, and 60°C) are pre-stored. The OCV-SOC mapping relationship at the current temperature is generated using the cubic spline interpolation method based on the battery cell temperature T collected by the BMS: OCV cal(T) = a·T 3 + b·T 2 + c·T + d; Among them, the coefficients a, b, c, and d are determined by fitting the reference curve data of adjacent temperature points; Aging-adaptive OCV correction: Dynamically update the OCV-SOC mapping table based on the battery state of health (SOH): When SOH ≥ 90% (battery health), the original baseline curve is used; When 80%≤SOH<90% (mild aging), the OCV value is corrected for negative offset (offset = 0.05V×(90%-SOH)); When SOH is less than 80% (severe aging), the OCV-SOC curve is refitted based on historical cycle data (updated every 24 hours) (the slope is reduced by 0.8 times and the intercept is shifted by 0.1V); OCV denoising under vibration interference: A 5-point sliding average filter is used to smooth the OCV measurement value (sampling frequency 5Hz). If the voltage fluctuation at a single sampling point is greater than 50mV (determined to be a transient interference caused by ship vibration), the point is eliminated and replaced by linear interpolation of the two points before and after. When the standard deviation of the filtered OCV value is ≤10mV, it is considered a valid measurement value. Multi-stage correction weight distribution: Dynamically adjust the weight coefficient α of the ampere-hour integral and OCV correction according to the battery operation stage (0≤α≤1): At the end of charging (current ≤ 0.05C and lasting 1 minute): α = 0.8 (OCV correction value is preferred); End of discharge (current ≤ 0.05C and lasting 1 minute): α = 0.7 (combined with OCV and integral value); Stable cruise phase (current 0.2C~0.5C): α=0.3 (mainly based on ampere-hour integration, with OCV auxiliary correction); SOC error closed-loop correction: Calculate the state of charge (SOC) by integrating ampere-hours amp = SOC initial + (∫I·ηdt) / C n ;SOC amp State of charge calculated by ampere-hour integration, SOC initial is the initial state of charge, I is the battery charge and discharge current; η is the Coulomb efficiency, η=0.98 when charging, η=0.95 when discharging; dt is the time element (integral time interval), C n is the rated capacity of the battery; OCV correction of state of charge SOCocv = f(OCV cal (T), SOH); SOC ocv is the state of charge after OCV correction, f(˙) is the mapping function between OCV and SOC, OCV cal (T) is the open circuit voltage after temperature compensation, T is the battery cell temperature, and SOH is the battery health status; Final SOC = α·SOC ocv + (1-α)·SOC amp .

[0063] What needs to be further explained in the embodiments of the present invention is that the above-mentioned calculation process is aimed at complex scenarios such as the dynamic operating conditions, wide temperature range, aging characteristics and vibration interference of ship batteries. Through multi-method fusion and parameter adaptive adjustment, high-precision real-time calculation of SOC is achieved, which provides a reliable basis for energy optimization scheduling (such as coordinated power supply of lithium batteries and supercapacitors) and active safety protection (such as over-discharge protection) of the energy control system.

[0064] Background: In ship energy control systems, battery state of health (SOH) is a core indicator for measuring the degree of battery performance degradation. It directly affects energy scheduling strategies (such as limiting the charge and discharge power of aging batteries), life prediction (determining replacement cycles), and safety protection (avoiding malfunctions caused by aging battery failure). However, the operating environment of ship batteries is complex, and existing single assessment methods have the following significant limitations: The most direct indicator of aging is the SOH, which is assessed by measuring the ratio of the battery's actual capacity to its initial capacity. However, this requires the battery to be fully charged and discharged (taking several hours), making it difficult to obtain in real time during dynamic ship operation. Aging is assessed by measuring changes in the battery's AC impedance (increases in internal resistance reflect aging phenomena such as electrode polarization and electrolyte attenuation). This can be measured online (by injecting a small current signal), but the internal resistance does not change significantly during initial aging (increases <5% when SOH ≥ 90%), resulting in insufficient assessment accuracy.

[0065] In addition, a single method is difficult to cover multi-dimensional aging characteristics (such as capacity decay reflects the loss of active materials, and internal resistance increase reflects ion transport barriers), so multiple methods are needed to complement each other. At the same time, battery aging can be divided into multiple stages, and the weights of the evaluation methods need to be adjusted at different stages to avoid over-reliance on internal resistance (large errors) in the early stage or ignoring internal resistance (a key indicator) during severe aging. Based on this: In a possible embodiment, the comprehensive assessment of health status includes: Internal resistance increment analysis using multi-frequency AC impedance method: Select characteristic frequency points to inject sinusoidal current signals, synchronously collect the battery terminal voltage response, and calculate the AC impedance at each frequency point through fast Fourier transform (FFT); Calculate the incremental rate of AC impedance at each frequency point relative to the initial value, and construct an internal resistance aging correlation model based on the incremental rate and weight coefficient; Capacity fading model of temperature stress-cycle number coupling: The temperature acceleration factor is calculated based on the Arrhenius equation, and the equivalent full charge and discharge cycle number is obtained using the acceleration factor and the actual cycle number; A capacity decay model is constructed based on the number of equivalent full charge and discharge cycles, initial capacity, and decay constant. Multi-stage dynamic weight fusion evaluation: Dynamically adjust the weight coefficient β (0≤β≤1) of the internal resistance aging correlation model and the capacity attenuation model according to the current SOH value: When SOH ≥ 90% (initial aging): β = 0.3 (capacity decay model dominates, internal resistance change is not significant); When 80%≤SOH<90% (mid-term aging): β=0.6 (weight balance between internal resistance increment and capacity model); When SOH is less than 80% (severe aging): β = 0.8 (internal resistance increment dominates, capacity decay rate slows down but internal resistance increases significantly); Final SOH = β × SOH imp + (1-β)×SOH cap ; Among them, SOH imp State of health (SOH) estimated for the internal resistance aging correlation model, SOH cap State of health (SOH) estimated for capacity fade model.

[0066] What needs to be further explained in the embodiments of the present invention is that the above-mentioned comprehensive evaluation process injects a sinusoidal current signal (covering high frequency to low frequency), calculates the impedance increment at each frequency point through FFT (reflecting different aging mechanisms: high-frequency impedance reflects ohmic internal resistance, medium-frequency impedance reflects charge transfer impedance, and low-frequency impedance reflects diffusion impedance), constructs an internal resistance aging correlation model, and realizes online and rapid evaluation; calculates the temperature acceleration factor based on the Arrhenius equation (quantifying the promoting effect of temperature on aging), and combines the actual number of cycles to obtain the equivalent full charge and discharge number, and constructs a capacity attenuation model to reflect the long-term cumulative effect of aging; adjusts the weights of the two methods according to the current SOH value: in the initial stage, the capacity model is mainly used, the weight is balanced in the medium term, and the internal resistance model is mainly used in severe aging to match the dominant aging mechanisms at different stages; the above-mentioned technical features realize high-precision and real-time evaluation of SOH under multi-factor aging, multi-stage heterogeneity and complexity of the operating environment of ship batteries.

[0067] In step S3 above, the dynamic energy optimization algorithm uses a model predictive control (MPC) framework with the objective function of minimizing system energy consumption and maximizing energy conversion efficiency. By inputting real-time data on main circuit current, key component temperatures, battery cell state of charge (SOC), state of health (SOH), and bus voltage, it calculates the optimal energy allocation ratio for each energy module within the optimization cycle and generates corresponding MPC instructions (such as the duty cycle adjustment value of the DC-DC converter and the output power target of the inverter). An isolated bidirectional DC-DC converter performs bidirectional energy conversion between the energy module and the bus, and adjusts the duty cycle according to MPC instructions to achieve bus voltage fluctuation suppression (≤±5%). A grid-connected DC-AC inverter converts the bus DC energy into usable AC load energy, and adjusts the output power according to MPC instructions. After executing the MPC instruction, the bus voltage and output power are collected in real time and fed back to the model predictive control (MPC) framework to correct the optimization calculation of the next cycle.

[0068] Background: In ship energy control systems, energy scheduling must take into account complex requirements such as dynamic load response, multi-objective optimization, equipment safety, and multi-energy coordination. Traditional fixed parameter control methods (such as PID control) have the following drawbacks: The load is highly dynamic and the traditional control response is delayed: The ship's propulsion load is affected by factors such as speed, propeller speed, and draft depth, and exhibits strong nonlinear and sudden changes. Traditional PID control relies on current error adjustment output and lacks prediction of future load changes, which can easily lead to excessive bus voltage fluctuations or unbalanced power distribution of energy modules.

[0069] Multi-objective optimization demand contradiction: Energy scheduling needs to minimize system energy consumption and maximize energy conversion efficiency at the same time, but the priorities of the two conflict under different working conditions: The startup condition requires a quick response (prioritizing efficiency), and excessive energy conservation may lead to insufficient power; Cruising conditions require balanced optimization (giving equal emphasis to energy consumption and efficiency); During port operation, the energy consumption of auxiliary equipment needs to be reduced (focusing on energy saving); Fixed-weight objective functions cannot dynamically balance these contradictions.

[0070] Equipment life and safety constraints are strict: Battery aging (decreased SOH) will cause the charging and discharging capabilities to decline. If its power upper limit is not limited, it may accelerate aging or cause overcharging and over-discharging. The overheating risk of key components such as the high-voltage contactor 4 contacts (≤90℃) and the current-limiting resistor 1 (≤150℃) needs to be avoided in advance. Traditional control relies only on real-time threshold triggering protection and lacks predictive constraints on future temperatures.

[0071] Multi-energy modules have very different characteristics: The energy control system provided by this invention integrates energy modules such as lithium batteries (high energy density, suitable for continuous power supply), supercapacitors (high power density, suitable for instantaneous response), and diesel generators (backup, high noise). Their power characteristics, response speeds, and operating costs vary significantly. Traditional control systems have difficulty dynamically adjusting module priorities based on operating conditions (for example, giving priority to supercapacitors during rapid acceleration), resulting in low energy utilization efficiency. Based on this: In one possible embodiment, a model predictive control (MPC) framework includes: Dynamic load prediction model: Based on the ship's real-time navigation data (speed v, propeller speed n, draft d) and historical operation database, the recursive least squares method is used to establish a propulsion power prediction model: P load (k+τ) = a·v(k+τ) 3 + b·n(k+τ) 2 + c·d(k+τ) + d0; Among them, P load (k+τ) is the predicted value of propulsion load power in the τth time step in the future, k is the current discrete time step (reference time), τ is the prediction time domain step, v(k+τ) is the predicted value of ship speed in the τth time step in the future, n(k+τ) is the predicted value of propeller speed in the τth time step in the future, d(k+τ) is the predicted value of ship draft in the τth time step in the future, a, b, c, d0 are the ship hydrodynamic coefficients fitted based on the navigation data of the past 30 days; Among them, the prediction time domain length T p Dynamic adjustment according to load change rate: When |dP load When / dt|>50kW / s (load sudden change), T p =3×100ms=300ms; when |dP load When / dt|≤50kW / s (stable working condition), T p =5×100ms=500ms); Multi-objective weight dynamic allocation strategy: According to the ship's operating conditions (starting / cruising / docking), the weight coefficients ω1 and ω2 (ω1+ω2=1) of minimizing system energy consumption (J1) and maximizing energy conversion efficiency (J2) in the objective function are dynamically adjusted: Startup conditions (speed v≤5 knots, duration≤2 minutes): Prioritize quick response, ω1=0.3 (energy consumption weight), ω2=0.7 (efficiency weight), to avoid power shortage due to excessive pursuit of energy saving; Cruising condition (5 knots < v ≤ 12 knots, lasting ≥ 10 minutes): balanced optimization, ω1 = 0.5, ω2 = 0.5, taking into account both long-term energy consumption and real-time efficiency; Port operating conditions (v≤3 knots, duration≤5 minutes): focus on energy saving, ω1=0.7, ω2=0.3, and reduce the energy consumption of auxiliary equipment; The objective function J = ω1˙J1 + ω2˙J2, where: J1 = Σ(τ = 1 to T p ) [P loss (τ)](P loss is the sum of conversion losses of each energy module); J2 = Σ(τ = 1 to T p ) [η avg (τ) / η max ](η avg is the average efficiency, η max is the maximum efficiency of the module); Lifetime constraints and temperature safety boundaries are embedded: Battery charge and discharge power constraint: P batt (τ)≤P max ×(1 - (100% - SOH) / 100%), P batt (τ) is the battery charge and discharge power at the next τth time step (charging is positive, discharging is negative), P max The maximum allowable charge and discharge power of the battery in a healthy state (SOH = 100%), SOH is the battery health state (the lower the SOH, the lower the power limit, delaying aging; for example, when SOH = 80%, P max reduced to 80% of its original value); Temperature safety constraint: T contactor (τ)≤90℃,T resistor (τ)≤150℃; T contactor (τ) is the contact temperature of high-voltage contactor 4 in the next τth time step, T resistor (τ) is the surface temperature of the current limiting resistor 1 at the next τth time step; Bus voltage strong constraint: ΔV bus (τ)≤±5%;ΔV bus (τ) is the voltage fluctuation amplitude of the high-voltage DC bus at the next τ-th time step; Collaborative optimization of multiple energy modules: Based on the differences in characteristics of lithium batteries (high energy density), supercapacitors (high power density), and diesel generators (backup), a module priority matrix was designed: Cruising mode: lithium battery (60%) + supercapacitor (30%) + diesel generator (10%, enabled only when SOC < 20%); Starting / rapid acceleration conditions: supercapacitor (50%) + lithium battery (40%) + diesel generator (10%), taking advantage of the supercapacitor's fast response characteristics; Port operation: lithium battery (70%) + supercapacitor (20%), diesel generator shut down to reduce noise; During optimization, the Lagrange multiplier method is used to solve the allocation ratio of each module to ensure the total output power P total (τ)=P load (τ)+ΔP buffer (τ); P total (τ) is the total output power at the next τth time step, P load (τ) is the load power demand in the next τ-th time step, ΔP buffer (τ) is the busbar capacitor buffer power; Online model self-update: After every 10 optimization cycles, the deviation between the actual output power and the predicted load is calculated using ΔP = P actual -P predicted ,The hydrodynamic coefficients a, b, c, and d0 are updated through Kalman filtering, so that the prediction model can adapt to changes in ship load (e.g., reduced draft after unloading) or sea conditions (e.g., increased wind and waves), and avoid long-term operation error accumulation.

[0072] In this embodiment of the present invention, it is important to further explain that the aforementioned model predictive control (MPC) framework constructs a load prediction model using the recursive least squares method. This model, combined with the vessel's real-time navigation data, predicts future loads, enabling control instructions to proactively adapt to load changes and prevent excessive bus voltage fluctuations. The weights of energy consumption (J1) and efficiency (J2) are adjusted based on the vessel's operating conditions (startup, cruising, and docking) to achieve optimal control under different operating conditions. Dynamically limiting charge and discharge power based on the SOH delays aging; limiting the future temperatures of contactor contacts and current-limiting resistor 1 to prevent overtemperature failures; and implementing strong bus voltage constraints to forcibly limit voltage fluctuations and ensure load power supply stability. Based on the differences in the characteristics of lithium batteries, supercapacitors, and diesel generators, a priority matrix related to the operating conditions is designed. The optimal allocation ratio is determined using the Lagrange multiplier method, ensuring that the total output power matches the load demand. Furthermore, the prediction model automatically adapts to changes in vessel load or sea conditions to prevent long-term error accumulation. The above technical features achieve the foresight, adaptability, safety and efficiency of energy scheduling through a closed-loop mechanism of "prediction-optimization-feedback", providing technical support for the stable operation and performance optimization of ship energy control systems.

[0073] In step S4 above, when an overcurrent, overvoltage or overtemperature anomaly is detected, the intelligent control module sends a fault signal to the high-voltage contactor 4 control unit and the fuse 5 monitoring unit synchronously via the CAN bus (500kbps rate), preferentially triggering the high-voltage contactor 4 to disconnect and simultaneously starting the fuse 5 melting countdown (5ms) to ensure that the fault energy is limited to a safe range; When an anomaly is triggered, the intelligent control module records the fault type (overcurrent / overvoltage / overtemperature), occurrence time, and peak parameters (such as maximum current 900A and maximum temperature 160°C) in real time, and uploads this information to the upper-level energy management system (EMS) via the Ethernet interface (100Mbps) for subsequent fault tracing and system optimization. Among them, a high-voltage contactor 4 supporting 1000V DC voltage is used. After receiving the fault signal sent by the intelligent control module, the main contact disconnection time is ≤10ms, ensuring that the main circuit is cut off before the fault current rises to the peak value; Configure a fast-acting fuse with a breaking capacity of ≥20kA. Its time-current characteristic matches the breaking time of high-voltage contactor 4. If high-voltage contactor 4 fails to break (for example, contact adhesion), fuse 5 will melt within 5ms to achieve backup isolation. If high-voltage contactor 4 successfully breaks, fuse 5 will remain inactive to avoid unnecessary replacement. Abnormal detection conditions: Overcurrent abnormality: The main circuit current value collected in real time by the current sensor exceeds 150% of the rated current of the high-voltage main circuit, and continues to exceed this threshold for three consecutive sampling cycles; Overvoltage abnormality: The 1000V high-voltage DC bus voltage monitored by the intelligent control module exceeds +5% of the rated value, and the abnormal state lasts for ≥20ms; Overtemperature anomaly: The contact temperature of high-voltage contactor 4, collected by the temperature sensor, is ≥120°C, the core temperature of fuse 5 is ≥150°C, or the surface temperature of pre-charge circuit current-limiting resistor 1 is ≥180°C (triggered at any position), and exceeds the corresponding threshold for two consecutive sampling cycles.

[0074] In step S1 above, when the system starts, it scans the connected energy modules through broadcast frames (CAN ID: 0x001) and receives the device descriptors (including energy type, rated power, voltage range, and communication protocol priority) fed back by the energy modules. The intelligent control module detects the energy module's voltage range (must cover the bus voltage), communication protocol compatibility (support at least one system built-in protocol), and protection level (≥IP65). If a match fails, the module will be denied access and a "module incompatibility" fault will be reported. After the matching is successful, a unique device ID is assigned to the energy module (such as 0x0101 for lithium battery pack No. 1), and the physical connection with the high-voltage DC bus is completed by closing auxiliary contactor 2.

[0075] In step S1 above, when the system starts, the main circuit high-voltage contactor 4 is in the disconnected state or the initial voltage of the bus capacitor is ≤ 10% of the rated voltage. The intelligent control module detects the "start pre-charge" instruction sent by the upper energy management system (EMS) and triggers the pre-charge process: The intelligent control module sends a closing command to the auxiliary contactor 2, which closes within 5ms and connects the current-limiting resistor 1 in series to the main circuit. The bus capacitor starts to charge through the current limiting resistor 1, and the control logic circuit 3 monitors the voltage across the capacitor in real time with a sampling frequency of 100Hz; When the capacitor voltage reaches 90% of the bus rated voltage or the pre-charge time reaches 500ms (whichever is triggered first), the control logic circuit 3 determines that the pre-charge is completed; The intelligent control module sends a closing command to the main circuit high-voltage contactor 4 (main contactor pull-in time ≤ 15ms), and simultaneously sends an opening command to the auxiliary contactor 2 (auxiliary contactor 2 breaking time ≤ 8ms), completing the main circuit switching; Among them, if the capacitor voltage rise rate is detected to be lower than 0.5V / ms during the pre-charging process (possibly due to an open circuit in the current limiting resistor 1 or a capacitor failure), the control logic circuit 3 immediately disconnects the auxiliary contactor 2 and reports the "pre-charging failure" fault to the upper energy management system (EMS) to avoid surge shock caused by malfunction of the main circuit high-voltage contactor 4.

[0076] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An energy control system for ship applications, characterized in that: include: The high-voltage main circuit control module includes a high-voltage contactor, a fuse, and a pre-charging circuit. The high-voltage contactor is used for on-off control of the main circuit, the fuse is used for overcurrent protection, and the pre-charging circuit includes a current-limiting resistor, an auxiliary contactor, and a control logic circuit for pre-charging when the system starts. Data acquisition module, including current sensor, temperature sensor and battery management system interface, used to collect main circuit current, key component temperature and battery status data respectively; an intelligent control module, including a BMS module control and communication unit, configured to execute an energy optimization algorithm with minimizing system energy consumption and maximizing energy conversion efficiency as objective functions, and to interact with a higher-level energy management system; Expansion and protection module, including high-voltage connectors, busbars and metal boxes, for accessing multiple energy modules; a multi-stage power electronic conversion module electrically connected to the high-voltage DC bus and configured to adjust the duty cycle and output power based on the energy optimization result; The heat dissipation module is linked with the temperature sensor to dissipate heat from key components to maintain the system operating temperature.

2. An energy control method for ship applications based on the control system according to claim 1, characterized in that: The following steps are involved: Start the system, connect multiple energy modules through the expansion and protection module and communication protocol, and pre-charge the bus capacitor using the current-limiting resistor of the pre-charge circuit; The main circuit current data is monitored in real time through the current sensor, the temperature data of key components is collected through the temperature sensor, and the battery cell voltage, temperature, state of charge and health status data are obtained through the battery management system interface; Based on the collected data, the BMS module control and communication unit executes a dynamic energy optimization algorithm. Combined with the multi-stage power electronic conversion module, it dispatches energy for the 1000V high-voltage DC bus, coordinates charging and discharging strategies, distributes energy among different energy modules, and controls voltage fluctuations. When overcurrent, overvoltage or overtemperature anomalies are detected, the main circuit is quickly cut off through the high-voltage contactor, and the fuse is triggered to blow and isolate the fault according to the cut-off result.

3. The energy control method for ship applications according to claim 2, characterized in that: Real-time monitoring of current data includes: The current sensor is installed in series on the busbar of the high-voltage main circuit to collect current data in real time at a preset sampling frequency; The collected current data is transmitted to the intelligent control module in real time and processed using an adaptive sliding average filtering strategy that includes dynamic evaluation of noise levels, adaptive adjustment of filtering parameters, optimization of operating condition associations, and isolation of abnormal data. It is then used for dynamic calculation of the energy optimization algorithm and real-time determination of overcurrent faults.

4. The energy control method for ship applications according to claim 2, characterized in that: When collecting temperature data, include: The temperature sensors are respectively integrated on the surface of the high-voltage contactor contact, the fuse core and the current-limiting resistor of the pre-charge circuit, and the temperature data is collected in real time at a preset sampling frequency; The collected temperature data is transmitted to the intelligent control module in real time. After being processed by segmented linearization correction and ambient temperature compensation, it is used to trigger the start-up of the heat dissipation module and serve as the basis for determining over-temperature faults. Among them, the piecewise linearization correction and ambient temperature compensation processing include: piecewise linearization correction of nonlinear characteristics, dynamic compensation of ambient temperature field and self-correction of thermal resistance aging.

5. The energy control method for ship applications according to claim 2, characterized in that: When obtaining battery cell data, including: The battery cell voltage is measured through the BMS's built-in 16-bit high-precision analog-to-digital converter; The battery cell temperature is collected through the NTC thermistor sensor integrated in the BMS; The state of charge is calculated by the ampere-hour integration method combined with open circuit voltage correction; Comprehensively evaluate the health status through battery internal resistance increment analysis and capacity decay model; The acquired battery cell voltage, temperature, state of charge, and health status data are transmitted to the intelligent control module in real time through the battery management system interface. After CRC verification and outlier filtering, they are used in the energy optimization algorithm. Among them, the calculation of the state of charge includes: dynamic stable operating condition identification and correction triggering, temperature compensated OCV-SOC mapping table construction, aging-adaptive OCV correction, OCV denoising processing under vibration interference, multi-stage correction weight allocation and SOC error closed-loop correction.

6. The energy control method for ship applications according to claim 5, characterized in that: A comprehensive assessment of health status includes: Select characteristic frequency points to inject sinusoidal current signals, synchronously collect the battery terminal voltage response, and calculate the AC impedance at each frequency point through fast Fourier transform; Calculate the incremental rate of AC impedance at each frequency point relative to the initial value, and construct an internal resistance aging correlation model based on the incremental rate and weight coefficient; The temperature acceleration factor is calculated based on the Arrhenius equation, and the equivalent full charge and discharge cycle number is obtained using the acceleration factor and the actual cycle number; A capacity decay model is constructed based on the number of equivalent full charge and discharge cycles, initial capacity, and decay constant. The weight coefficients of the internal resistance aging association model and the capacity attenuation model are dynamically adjusted according to the current SOH value, and the final health status is output.

7. The energy control method for ship applications according to claim 2, characterized in that: The dynamic energy optimization algorithm uses a model predictive control framework to calculate the optimal energy allocation ratio for each energy module within the optimization cycle by inputting real-time main circuit current, key component temperature, battery cell state of charge, health status and bus voltage data, and generates corresponding MPC instructions; The isolated bidirectional DC-DC converter performs bidirectional energy conversion between the energy module and the bus, and adjusts the duty cycle according to the MPC instructions; The grid-connected DC-AC inverter converts the bus DC energy into AC load-usable electrical energy and adjusts the output power according to the MPC instructions; Among them, the model predictive control framework includes: a dynamic load prediction model that dynamically adjusts the prediction time domain length according to the load change rate, a multi-objective weight dynamic allocation strategy, life constraint and temperature safety boundary embedding, multi-energy module collaborative optimization, and online model self-update.

8. The energy control method for ship applications according to claim 2, characterized in that: When overcurrent, overvoltage or overtemperature anomalies are detected, the high-voltage contactor is triggered to disconnect first, and the fuse blow countdown is started at the same time; If the high-voltage contactor fails to disconnect, the fuse will melt within the set time to achieve backup isolation. If the high-voltage contactor successfully disconnects, the fuse will remain inactive.

9. The energy control method for ship applications according to claim 2, characterized in that: When the system starts, it scans the connected energy modules through broadcast frames and receives the device descriptors fed back by the energy modules; The intelligent control module detects the voltage range, communication protocol compatibility and protection level of the energy module; After the matching is successful, a unique device ID is assigned to the energy module, and the physical connection with the high-voltage DC bus is completed by closing the auxiliary contactor.

10. The energy control method for ship applications according to claim 2, characterized in that: When the system starts, the auxiliary contactor is closed and the current limiting resistor is connected in series to the main circuit; The bus capacitor begins to charge through the current-limiting resistor, and the control logic circuit monitors the voltage across the capacitor in real time; When the capacitor voltage or pre-charge time reaches the set value, the control logic circuit determines that the pre-charge is completed, closes the main circuit high-voltage contactor, and opens the auxiliary contactor at the same time to complete the main circuit switching.

Images