Energy-saving control method for electric ship container

Abstract

The invention discloses an energy-saving control method for an electric ship container, and relates to the technical field of ship container energy conservation, and the method comprises the following steps: monitoring the voltage, current and temperature of a battery pack in the ship container in real time through installing a battery monitoring device, and collecting data at a preset frequency to form a battery state data set; the temperature control strategy is dynamically adjusted by monitoring the battery temperature, the charging and discharging state and the environmental condition in real time and combining the precise cooling and heating system, the battery is prevented from being overheated or supercooled, the safety is improved, and the service life is prolonged. Meanwhile, the energy efficiency is intelligently optimized, the energy waste is reduced, and the operation cost is reduced. Machine learning and optimization algorithms are introduced, battery temperature changes are accurately predicted, preventive measures are taken, energy efficiency is improved, and green and sustainable operation of ships is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving for ship containers, and particularly to an energy-saving control method for electric ship containers. Background Art

[0002] The energy-saving control of electric ship containers refers to achieving efficient utilization and minimizing consumption of energy during the transportation of electric ships through an intelligent and optimized control system. Specifically, the energy-saving control system can dynamically adjust the operating mode and power output of the electric ship according to various factors such as the ship's load, speed, voyage, sea conditions, and weather. For example, the system can monitor the battery power in real time, optimize power distribution, avoid unnecessary energy waste, and at the same time optimize the charging and discharging strategies according to the voyage plan, and even reduce the energy consumption during navigation through intelligent route planning. In addition, energy-saving control also includes efficient power management of the ship, such as using regenerative braking technology to recover energy, intelligently regulating the operation of the electric propulsion system and auxiliary power equipment, so that the energy utilization of the ship reaches the best state, thereby improving energy efficiency, reducing operating costs, and reducing the negative impact on the environment.

[0003] The prior art has the following deficiencies:

[0004] In the energy-saving control process of existing electric ship containers, the energy management system (EMS) ignores the battery temperature fluctuations. During the charging and discharging process of the battery, heat is generated, especially in the case of high load or long-term operation, the battery temperature is likely to be too high or too low. If the energy-saving control system fails to effectively monitor and adjust the battery temperature, it may cause the battery to overheat or cool down, which in turn affects the cycle life and safety of the battery. For example, overheating can cause instability in the internal chemical reactions of the battery, increasing the risk of fire and even causing the battery to catch fire and explode; while overcooling may lead to a significant decrease in battery efficiency or even damage the battery. Long-term neglect of this problem may lead to frequent failures of the ship's battery system, increasing maintenance costs, and seriously threatening the safety of crew and cargo. Therefore, a reasonable temperature control and battery management mechanism should be a core link in the energy-saving control system of electric ships.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to provide an energy-saving control method for an electric ship container. By monitoring the battery pack temperature, charge and discharge status, and external environmental conditions in real time, and combining the precise adjustment of the cooling and heating systems, the system can dynamically adjust the temperature control strategy to avoid overheating or overcooling of the battery, improve the battery safety, stability, and extend its service life. At the same time, the temperature control system intelligently optimizes the energy efficiency according to the battery load and environmental changes, reduces energy waste, and lowers the operating cost. The introduction of machine learning and optimized scheduling algorithms makes the temperature control more accurate, predicts the battery temperature changes, and takes preventive measures, thereby improving the energy efficiency, reducing unnecessary energy consumption, and promoting the green and sustainable operation of the ship to solve the problems in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: An energy-saving control method for an electric ship container, comprising the following steps:

[0008] By installing a battery monitoring device, the voltage, current, and temperature of the battery pack in the ship container are monitored in real time, and data is collected at a preset frequency to form a battery status data set;

[0009] Through an environmental sensor, the external environmental temperature and humidity during the ship operation are collected in real time to generate an environmental data set;

[0010] Based on the temperature data in the battery status data set, when the battery temperature exceeds the preset safety threshold, the cooling system is started for cooling; when the battery temperature is lower than the preset threshold, the heating system is started for heating to keep the battery within the ideal operating temperature range;

[0011] Based on an optimized scheduling algorithm for minimizing energy consumption, according to the real-time power, charge and discharge status of the battery pack, and environmental data, the charge and discharge strategy of the battery pack is dynamically adjusted to achieve the optimal energy use efficiency;

[0012] Based on the trained machine learning model, the future change trend of the battery temperature is predicted, and the operation strategy of the cooling or heating equipment is adjusted in advance to ensure that the battery is always within the safe temperature range.

[0013] Preferably, the specific steps of installing a battery monitoring device to monitor the voltage, current, and temperature of the battery pack in the ship container in real time and collecting data at a preset frequency to form a battery status data set are as follows:

[0014] First, connect the battery monitoring device to the electrical interface of the battery pack to obtain the voltage, current, and temperature data of the battery pack in real time;

[0015] The battery monitoring device processes the real-time collected battery voltage, current, and temperature data according to the preset collection frequency;

[0016] The data transmitted to the control system in real time will be processed by the data processing module to form a time-sensitive battery status dataset;

[0017] The processed battery status data will be stored in the central control system or cloud server to form a battery health monitoring database.

[0018] Preferably, the specific steps for generating the environmental dataset by collecting the external environmental temperature and humidity during the ship operation in real time through the environmental sensor are as follows:

[0019] While the battery monitoring device monitors the battery status in real time, the environmental sensor starts to collect the external environmental temperature and humidity data during the ship operation in real time;

[0020] The temperature and humidity data collected by the environmental sensor will be collected and transmitted to the central control system at a preset frequency;

[0021] Once the environmental data is collected and transmitted to the control system, it will be processed and analyzed together with the battery status data;

[0022] The processed environmental data will be stored in the database of the central control system to form a complete environmental dataset.

[0023] Preferably, based on the temperature data in the battery status dataset, when the battery temperature exceeds the preset safety threshold, the cooling system is started for cooling; when the battery temperature is below the preset threshold, the heating system is started for heating to keep the battery within the ideal operating temperature range. The specific steps are as follows:

[0024] The central control system compares the temperature data in the battery status dataset with the preset safety threshold in real time;

[0025] Once it is determined that the battery temperature exceeds the safe range, the control system will automatically adjust the working state of the cooling or heating equipment according to the temperature change;

[0026] After the cooling or heating system is started, the central control system will continue to monitor the change of the battery temperature and adjust the temperature control strategy in real time.

[0027] Preferably, based on the optimization scheduling algorithm for minimizing energy consumption, according to the real-time battery level, charge and discharge status of the battery pack and the environmental data, the charge and discharge strategy of the battery pack is dynamically adjusted to achieve the optimal energy use efficiency. The specific steps are as follows:

[0028] First, it is necessary to obtain the real-time battery level and charge and discharge status of the battery pack. These data are the basis for optimizing the charge and discharge strategy. The calculation expressions are as follows:

[0029]

[0030] , where E real is the current actual power of the battery pack, and I battery (t) is the instantaneous current of the battery, that is, the instantaneous current at time point t;

[0031] The charge and discharge state C status is determined by monitoring the direction and intensity of the battery current. If the current is positive, the battery is discharging; if the current is negative, the battery is charging;

[0032] When formulating the charge and discharge strategy, environmental data has a direct impact on the energy efficiency and performance of the battery. Therefore, environmental data is taken into consideration, and correction factors are set according to the influence of temperature and humidity on the battery performance to adjust the charge and discharge efficiency of the battery. The calculation expression is as follows:

[0033] f temp = 1 + α temp (T env - T opt ),

[0035] f hum = 1 + β hum (H env - H opt ),

[0036] In the formula, f temp is the environmental temperature correction factor, T env is the external environmental temperature, T opt is the optimal operating temperature of the battery, α temp is the influence coefficient of temperature on the charge and discharge efficiency of the battery, f hum is the environmental humidity correction factor, H env is the external environmental humidity, H opt is the optimal operating humidity of the battery, β hum is the influence coefficient of humidity on the charge and discharge efficiency of the battery,

[0037] Based on the current actual power E real of the battery pack, the charge and discharge state C status , as well as the environmental temperature correction factor f temp and the environmental humidity correction factor f hum , the charge and discharge power distribution is dynamically adjusted through an optimized scheduling algorithm. The calculation expression is as follows:

[0038]

[0039] , where L energy is the energy loss of the battery, E battery is the total power of the battery, P chargingP(t) is the charging power of the battery at time point t;

[0040] To further improve the energy usage efficiency, the optimized scheduling algorithm also needs to adjust the charging and discharging strategies by predicting the future change trend of the battery power. Use machine learning models or time series prediction methods to model the change trend of the battery power, predict the battery power change in the future time, and the calculation expression is as follows:

[0041]

[0042] , where, E predict (t + Δt) is the predicted battery power, that is, the predicted value of the battery power within the time interval Δt, and E real (t) is the actual battery power at time point t, and P net (t) is the battery network power, the net power of the battery at time point t.

[0043] Preferably, based on the trained machine learning model, predict the future change trend of the battery temperature, and adjust the operation strategy of the cooling or heating equipment in advance to ensure that the battery is always within the safe temperature range. The specific steps are as follows:

[0044] To train the machine learning model and perform temperature prediction, first, a large amount of historical battery temperature data needs to be collected. These data will be used as input features. The real-time data collected by the battery monitoring device and the environmental sensor will form a data set and be input into the machine learning model. To ensure the accuracy and availability of the data, the data needs to be preprocessed before training. Let the collected battery temperature data be T(t), the environmental temperature data be E(t), and the battery voltage and current data be V(t) and I(t) respectively, where t represents the time point;

[0045] The data set after data preprocessing will form the basis for model training, ensuring that the model can more accurately learn the law of temperature change. For each time point t, the historical data of temperature change will be constructed into a feature set X(t) = [T(t - k), E(t - k), V(t - k), I(t - k)], where k is the size of the sliding window, representing the range of historical data used for prediction. T(t - k) is the historical battery temperature data, E(t - k) is the historical environmental temperature data, V(t - k) is the historical battery voltage data, and I(t - k) is the historical battery current data;

[0046] Use the preprocessed data set to train a machine learning model. By fitting the relationship between the battery temperature and its related features, predict the future change trend of the battery temperature. During the training process, optimize the parameters of the model by minimizing the loss function. The expression is as follows:

[0047]

[0048] , where L is the minimized loss function, T(t + k) is the true battery temperature, is the battery temperature predicted by the model, which is the battery temperature at the future time (t + k) predicted by the machine learning model based on the current input features, and N is the number of samples;

[0049] After the machine learning model is trained, it is used to predict the change trend of the battery temperature within the future time point t + k. By inputting the current state features X(t), the model will output the predicted value of the future temperature (t + k). Suppose it is necessary to predict the temperature for the next N steps in the future. The model will sequentially output the predicted temperatures at each time step

[0050] Output of the model is the temperature prediction based on the state features at the current moment, and its formula is as follows:

[0051]

[0052] , where f(X(t)) is the model function, representing the machine learning model, which maps the input feature set X(t) to the predicted value of the battery temperature ;

[0053] According to the temperature prediction result of the machine learning model, adjust the operation strategy of the cooling or heating equipment in advance. When it is predicted that the future battery temperature will exceed the upper threshold T max , the system will start the cooling equipment in advance; when it is predicted that the temperature is lower than the lower threshold T min , start the heating equipment. In order to ensure that the battery temperature always remains within the safe range, the control system makes dynamic adjustments according to the prediction result, using the following formula:

[0054]

[0055] , where ΔT adjust (t + k) is the adjustment amplitude of the battery temperature change, representing the difference between the predicted future temperature and the current actual temperature, and T current is the current actual temperature of the battery;

[0056] If Δ Tadjust (t + k)>0, it indicates that the future temperature is too high, and start the cooling equipment; if Δ Tadjust (t + k)<0, it indicates that the temperature is too low, and start the heating equipment. Further, the control system sets the power of the temperature control equipment according to the temperature adjustment amplitude Δ Tadjust to ensure that the temperature change is within the predetermined range. The specific temperature control strategy is adjusted through the following model:

[0057] P cooling = α × ΔT adjust (t + k),

[0059] P heating = β × |ΔT adjust (t + k)|,

[0061] where α and β are the adjustment coefficients of the cooling and heating systems, and P cooling is the power required to start the cooling system, and P heating is the power required to start the heating system.

[0062] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0063] By monitoring the temperature, charge-discharge state, and external environmental conditions of the battery pack in real time and combining with the precise adjustment of the cooling and heating systems, the present invention can significantly improve the safety and stability of the battery. Traditional battery management systems mainly rely on fixed temperature thresholds for control, but this method may suffer from response lags or insufficient temperature control. By adopting a dynamic temperature control scheme (such as a temperature control scheduling system based on battery monitoring devices, machine learning models, or comprehensive energy efficiency optimization), the system can judge the working state of the battery in real time and make precise temperature adjustments according to the actual needs of the battery and changes in the external environment. Timely start the cooling or heating equipment to avoid extreme situations of overheating or overcooling of the battery, thereby effectively preventing performance degradation, capacity decline, and even safety accidents caused by excessive temperature fluctuations of the battery.

[0064] At the same time, the temperature control system can adaptively adjust according to the charge-discharge load of the battery pack, environmental conditions, and the health state of the battery, reduce the amplitude of temperature fluctuations, slow down the battery aging process, and extend the service life of the battery. By optimizing temperature management, the chemical reactions of the battery and the working environment of the internal components of the battery are better protected, avoiding accelerated chemical reactions caused by high temperatures or a decrease in battery efficiency caused by low temperatures. This intelligent temperature control strategy not only extends the battery life cycle but also greatly reduces battery failures and maintenance costs caused by improper temperature management.

[0065] Based on the combination of a temperature control scheduling system and comprehensive energy efficiency optimization, the ship can utilize energy more efficiently during operation, maximize the energy efficiency of the battery, and thereby reduce the overall operating cost. During the ship's voyage, the system dynamically adjusts the working intensity of the temperature control system according to the battery load, charge and discharge status, and external environmental changes, avoiding excessive energy consumption. This not only reduces the energy waste of cooling or heating equipment but also avoids the energy efficiency loss caused by too high or too low battery temperature. For example, when the battery load is light, the system will reduce the power of cooling or heating to save energy; while in the case of high load or low battery power, the system will appropriately increase the power of the temperature control equipment to ensure that the battery operates in the best working state.

[0066] In addition, the introduction of machine learning and optimized scheduling algorithms makes the temperature control system more intelligent and precise, capable of predicting battery temperature changes in advance and taking preventive measures to reduce energy waste. Through the analysis of historical data and real-time feedback, the system can gradually optimize the temperature control strategy, improve energy utilization efficiency, and reduce unnecessary energy consumption. In the long run, this intelligent energy management not only improves the overall operating efficiency of the ship but also effectively reduces energy costs, reduces the negative impact on the environment, and promotes the green and sustainable operation of the ship. Brief Description of the Drawings

[0067] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0068] Figure 1 It is a flowchart of a method for energy-saving control of an electric ship container according to the present invention. Detailed Embodiments

[0069] Now, the exemplary embodiments will be described more comprehensively with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0070] The present invention provides an energy-saving control method for an electric ship container as shown in Figure 1 and includes the following steps:

[0071] By installing a battery monitoring device, the voltage, current, and temperature of the battery pack in the ship container are monitored in real time, and data is collected at a preset frequency to form a battery state data set;

[0072] By installing a battery monitoring device, the specific steps to monitor the voltage, current, and temperature of the battery pack in a ship container in real time and collect data at a preset frequency to form a battery status dataset are as follows:

[0073] First, connect the battery monitoring device to the electrical interface of the battery pack to obtain the voltage, current, and temperature data of the battery pack in real time;

[0074] The battery monitoring device usually includes multiple sensors and a data acquisition module, which can monitor the voltage change, charge and discharge current of the battery pack, and the temperature distribution of each battery cell. These data help to timely understand the working state of the battery, and can reflect the battery's health status, load changes, and whether there are potential fault risks. The measurement of battery voltage and current is carried out through precise voltage sensors and current sensors, while temperature sensors are arranged at key positions inside and outside the battery pack to comprehensively monitor temperature fluctuations.

[0075] The battery monitoring device processes the real-time acquired battery voltage, current, and temperature data according to the preset acquisition frequency;

[0076] The acquisition frequency can usually be set according to specific requirements, such as collecting data once per second, per minute, or per hour, to ensure real-time reflection of the operating state of the battery pack. The acquired data will be uploaded to the central control system or cloud platform through the data transmission module to ensure that the data can be effectively stored and analyzed. To improve the reliability of data transmission, wireless transmission technologies (such as Wi-Fi, Bluetooth, 4G / 5G, etc.) or wired network connections are usually used to transmit the data from the monitoring device to the remote processing system. The key to this step is to ensure the integrity and real-time nature of the data for subsequent processing and decision-making.

[0077] The data transmitted to the control system in real time will be processed by the data processing module to form a time-sensitive battery status dataset;

[0078] The data processing module preliminarily analyzes the real-time acquired battery voltage, current, and temperature data according to the set algorithm to evaluate the working state of the battery. For example, judge whether the battery is in a normal charging, discharging, or no-load state according to the battery voltage value, analyze the current data to evaluate whether the charge and discharge power of the battery is stable, and use the temperature data to judge whether the battery is overheated or overcooled. At this time, the system can also perform simple fault diagnosis, such as abnormal voltage, too high temperature, or too large current, and give real-time alarms and record relevant information. The data processing at this stage not only provides a basis for subsequent control decisions but also provides scientific support for the health management of the battery.

[0079] The processed battery status data will be stored in the central control system or cloud server to form a battery health monitoring database;

[0080] The data is not only used for real-time monitoring and analysis, but also provides data support for subsequent predictive maintenance. The stored data can be analyzed through historical data in the future to conduct trend prediction and evaluate the battery life and health status. During this process, the battery monitoring device also needs to be able to feedback information on battery state changes and send alerts to operators or control systems. For example, when the battery temperature exceeds the safe range or the voltage fluctuates abnormally, a fault warning is issued in a timely manner. This can avoid problems such as ship stoppage or unstable energy systems caused by battery failures.

[0081] The external environmental temperature and humidity during the ship's operation are collected in real time through environmental sensors to generate an environmental data set;

[0082] The specific steps for collecting the external environmental temperature and humidity during the ship's operation in real time through environmental sensors to generate an environmental data set are as follows:

[0083] While the battery monitoring device monitors the battery state in real time, the environmental sensors start to collect the external environmental temperature and humidity data during the ship's operation in real time;

[0084] The environmental sensors continuously detect the temperature and humidity changes in the ship's environment through precise temperature and humidity measurement devices. These environmental data are crucial for optimizing battery management and energy-saving control. Especially during navigation, temperature and humidity will directly affect the charging and discharging efficiency and life of the battery. For example, a high-temperature environment may cause the battery to overheat, while high humidity may accelerate the corrosion of external battery components. Therefore, the environmental data must be updated in real time to ensure that the system can adjust the control strategy in a timely manner according to the current external conditions.

[0085] The temperature and humidity data collected by the environmental sensors are collected and transmitted to the central control system at a preset frequency;

[0086] The data collection frequency is the same as that of the battery state data and can be set to once per second, per minute, or per hour according to requirements. The environmental data is transmitted to the central processing unit in real time through wireless or wired networks and forms a complete data set together with the battery state data. These data are not only the basis for the battery management system to adjust the operation mode, but also provide environmental basic data for subsequent energy efficiency optimization. During the data transmission process, the stability and real-time nature of communication must be ensured to prevent environmental changes from not being timely feedback to the control system.

[0087] Once the environmental data is collected and transmitted to the control system, it will be processed and analyzed together with the battery state data;

[0088] The data processing module in the control system will conduct a correlation analysis based on real-time environmental data (temperature, humidity) and battery status data (voltage, current, temperature). At this time, the environmental data provides important external background information for the operation of the battery. For example, the system will evaluate the current battery operation efficiency in combination with the environmental temperature data to determine whether it is necessary to adjust the cooling or heating equipment to cope with the change of external temperature. If the environmental humidity is too high, the system may activate the moisture-proof function in advance to protect the battery from corrosion. In this way, the battery management system can make more accurate control decisions, effectively improve energy efficiency and extend the battery life.

[0089] The processed environmental data will be stored in the database of the central control system to form a complete environmental data set;

[0090] In addition to real-time use, these historical data can also provide a long-term reference basis for future ship navigation and battery management. By analyzing the historical environmental data, the system can summarize the influence laws of different environmental conditions on battery performance and optimize future energy-saving control strategies. The system can also regularly review the past temperature and humidity data for trend prediction, plan in advance the impact of possible environmental changes on the battery, and make preventive adjustments. This not only helps the system to perform dynamic adjustment during ship operation, but also provides data support for the later ship maintenance and repair plan.

[0091] Based on the temperature data in the battery status data set, when the battery temperature exceeds the preset safety threshold, start the cooling system for temperature reduction; when the battery temperature is lower than the preset threshold, start the heating system for temperature increase to keep the battery within the ideal operating temperature range;

[0092] Based on the temperature data in the battery status data set, when the battery temperature exceeds the preset safety threshold, start the cooling system for temperature reduction; when the battery temperature is lower than the preset threshold, start the heating system for temperature increase to keep the battery within the ideal operating temperature range. The specific steps are as follows:

[0093] The central control system makes a real-time comparison between the temperature data in the battery status data set and the preset safety threshold;

[0094] The safety threshold is usually set according to the battery type, design parameters and usage conditions to ensure that the battery operates within a safe temperature range. At this time, the system will judge whether the battery temperature is higher or lower than the set safety range. If the battery temperature exceeds the upper limit threshold, it means that the battery may be in an overheated state, and the system will start the cooling system for temperature reduction; if the battery temperature is lower than the lower limit threshold, it indicates that the battery may face an overcooled situation, and the system will start the heating system for temperature increase. This comparison operation needs to be carried out in real time to ensure that the battery always maintains the best operating temperature.

[0095] Once it is determined that the battery temperature exceeds the safe range, the control system will automatically adjust the operating state of the cooling or heating equipment according to the temperature change;

[0096] When the battery temperature is too high, the system activates the cooling function through a cooling device (such as a liquid cooling system or an air cooling system) to quickly reduce the temperature to within the safe range. The cooling system usually includes components such as liquid or gas coolants and radiators, which carry away heat through active circulation. When the battery temperature is too low, the system will activate the heating system, for example, heating the battery through electric heating or hot air equipment to increase the battery temperature and avoid the impact of low temperature on battery performance. This process not only ensures that the battery operates within the safe temperature range but also avoids the negative impact of temperature fluctuations on battery health and lifespan.

[0097] After the cooling or heating system is activated, the central control system will continue to monitor the change of the battery temperature and adjust the temperature control strategy in real time;

[0098] Through comprehensive analysis of environmental data (such as external temperature and humidity) and battery temperature, the system can optimize the temperature control strategy. For example, in the case of a relatively low ambient temperature, the power of the heating system can be appropriately adjusted to avoid excessive temperature rise; in a high-temperature environment, the power of the cooling system may need to be further enhanced. At this time, the control system will also feedback the battery temperature data to the operator in real time to ensure that the ship management personnel can obtain timely status information. After the system runs for a period of time, the accumulation of historical data can help optimize the control algorithm, improve the temperature control strategy, and enhance the stability and efficiency of the entire battery management system.

[0099] Based on the optimization scheduling algorithm for minimizing energy consumption, according to the real-time power, charge and discharge status of the battery pack and environmental data, dynamically adjust the charge and discharge strategy of the battery pack to achieve the optimal energy use efficiency;

[0100] The specific steps of the optimization scheduling algorithm based on minimizing energy consumption to dynamically adjust the charge and discharge strategy of the battery pack according to the real-time power, charge and discharge status of the battery pack and environmental data to achieve the optimal energy use efficiency are as follows:

[0101] First, it is necessary to obtain the real-time power and charge and discharge status of the battery pack. These data are the basis for optimizing the charge and discharge strategy. The calculation expressions are as follows:

[0102]

[0103] , where E real is the current actual power of the battery pack, and I battery (t) is the instantaneous current of the battery, that is, the instantaneous current at time point t;

[0104] Charge and discharge status Cstatus By monitoring the direction and intensity of the battery current, it is determined that if the current is positive, the battery is discharging, and if the current is negative, the battery is charging;

[0105] The real-time battery power and charge / discharge status provide accurate basic data for the decision-making in the subsequent steps.

[0106] When formulating the charge / discharge strategy, the environmental data has a direct impact on the energy efficiency and performance of the battery. Therefore, the environmental data is taken into consideration, and a correction factor is set according to the influence of temperature and humidity on the battery performance to adjust the battery charge / discharge efficiency. The calculation expression is as follows:

[0107] f temp = 1 + α temp (T env - T opt ),

[0109] f hum = 1 + β hum (H env - H opt ),

[0110] In the formula, f temp is the environmental temperature correction factor, which takes into account the influence of the external temperature on the battery charge / discharge efficiency. T env is the external environmental temperature, and T opt is the optimal operating temperature of the battery. Different types of batteries (such as lithium batteries, lead-acid batteries, etc.) have different optimal operating temperature ranges, which are usually determined according to the specification sheets provided by the battery manufacturers. For lithium batteries, the optimal operating temperature is usually 20°C to 25°C. α temp is the influence coefficient of temperature on the battery charge / discharge efficiency. f hum is the environmental humidity correction factor, which is used to adjust the battery charge / discharge efficiency to cope with the battery operating state under different humidity conditions. H env is the external environmental humidity, and H opt is the optimal operating humidity of the battery. β hum is the influence coefficient of humidity on the battery charge / discharge efficiency.

[0111] The environmental temperature correction factor f temp is used to dynamically adjust the battery charge / discharge efficiency. If T env > T opt , then f temp will increase, indicating that too high temperature will cause the reduction of the battery charge / discharge efficiency; if T env < T opt , then f tempIt will be less than 1, indicating that low temperature will cause difficulties in battery charging or reduce the efficiency. Therefore, adjust the charge-discharge strategy to avoid the adverse effects of high or low temperature on battery performance.

[0112] Environmental humidity correction factor f hum Adjust the charge-discharge efficiency of the battery to cope with the battery operating state under different humidity conditions. When H env >H opt When it is the case, the humidity is high and f hum will increase, indicating that too high humidity will cause a decrease in battery efficiency, and the system may need to take dehumidification measures; when H env <H opt When it is the case, the humidity is low and f hum will be less than 1, indicating that low humidity will also have a negative impact on battery performance, and the charge-discharge strategy needs to be appropriately adjusted to improve battery efficiency.

[0113] Based on the current actual battery power E real , charge-discharge state C status , as well as the environmental temperature correction factor f temp and the environmental humidity correction factor f hum , dynamically adjust the charge-discharge power distribution through an optimized scheduling algorithm. The calculation expression is as follows:

[0114]

[0115] , where L energy is the energy loss of the battery, E battert is the total battery power, P charging (t) is the charging power of the battery at time point t;

[0116] The goal of this step is to minimize the total power of unit energy consumption while taking into account the influence of environmental factors. By dynamically adjusting the charging power, it can ensure that the energy consumption during the battery charge-discharge process is minimized, while avoiding overcharging or over-discharging.

[0117] To further improve the energy usage efficiency, the optimized scheduling algorithm also needs to adjust the charge-discharge strategy by predicting the future change trend of the battery power. Use a machine learning model or a time series prediction method to model the change trend of the battery power, predict the change of the battery power in the future time, and the calculation expression is as follows:

[0118]

[0119] , where E predict (t + Δt) is the predicted battery power, that is, the predicted value of the battery power within the time interval Δt, E real (t) is the actual battery power at time point t, Pnet (t) is the battery network power, the net power of the battery at time point t;

[0120] By predicting the future changes in battery power, the system can make charging and discharging decisions in advance. For example, it can increase the charging power when the battery power is expected to be insufficient, or reduce the charging power when the battery is expected to be full. The key to this step is to use accurate predictions of the battery power change trend to optimize the charging and discharging strategy and maximize the energy usage efficiency.

[0121] Based on the trained machine learning model, predict the future change trend of the battery temperature, and adjust the operation strategy of the cooling or heating equipment in advance to ensure that the battery is always within the safe temperature range;

[0122] Based on the trained machine learning model, predict the future change trend of the battery temperature, and adjust the operation strategy of the cooling or heating equipment in advance to ensure that the battery is always within the safe temperature range. The specific steps are as follows:

[0123] To train the machine learning model and perform temperature prediction, a large amount of historical battery temperature data needs to be collected first, including information such as the battery voltage, current, external environmental temperature, humidity, etc. These data will be used as input features. The real-time data collected by the battery monitoring device and environmental sensors will form a data set and be input into the machine learning model. To ensure the accuracy and availability of the data, data preprocessing is required before training, including removing outliers, filling missing values, standardization processing, etc. Let the collected battery temperature data be T(t), the environmental temperature data be E(t), and the battery voltage and current data be V(t) and I(t) respectively, where t represents the time point;

[0124] The data set after data preprocessing will form the basis for model training, ensuring that the model can more accurately learn the law of temperature change. For each time point t, the historical data of temperature change is constructed into a feature set X(t) = [T(t - k), E(t - k), V(t - k), I(t - k)], where k is the size of the sliding window, representing the range of historical data used for prediction. T(t - k) is the historical battery temperature data, which is the temperature data of the battery at the past time point t - k. E(t - k) is the historical environmental temperature data, which is the external environmental temperature data of the battery at the past time point t - k. V(t - k) is the historical battery voltage data, which is the voltage data of the battery at the past time point t - k. I(t - k) is the historical battery current data, which is the current data of the battery at the past time point t - k;

[0125] Train a machine learning model, such as a multi-layer perceptron (MLP), support vector regression (SVR), or long short-term memory network (LSTM), using the preprocessed dataset. By fitting the relationship between the battery temperature and its related features (such as battery voltage, current, ambient temperature, etc.), predict the future change trend of the battery temperature. During the training process, optimize the parameters of the model by minimizing the loss function, and the expression is as follows:

[0126]

[0127] , where L is the minimized loss function, T(t + k) is the true battery temperature, representing the actual temperature of the battery at time point t + k, is the battery temperature predicted by the model, which is the battery temperature at the future time (t + k) predicted by the machine learning model based on the current input features (such as battery voltage, current, ambient temperature, etc.). N is the number of samples, and each sample corresponds to a time point t and includes the comparison between the actual temperature T(t + k) and the predicted temperature ;

[0128] After the machine learning model is trained, it is used to predict the change trend of the battery temperature within the future time point t + k. By inputting the current state features X(t) (including battery voltage, current, ambient temperature, etc.), the model will output the predicted value of the future temperature Suppose it is necessary to predict the temperature for the next N steps. The model will sequentially output the predicted temperature for each time step

[0129] The output of the model is the temperature prediction based on the state features at the current moment, and its formula is as follows:

[0130]

[0131] , where f(X(t)) is the model function, representing the machine learning model, which maps the input feature set X(t) to the battery temperature prediction value ;

[0132] In this way, the system can not only predict the current temperature but also anticipate the future change trend of the temperature in advance, helping the decision-making system to make pre-cooling adjustments.

[0133] According to the temperature prediction results of the machine learning model, adjust the operation strategy of the cooling or heating equipment in advance. When it is predicted that the future battery temperature will exceed the upper threshold T max , the system will start the cooling equipment in advance; when the predicted temperature is lower than the lower threshold T min , start the heating equipment. To ensure that the battery temperature always remains within the safe range, the control system makes dynamic adjustments according to the prediction results, using the following formula:

[0134] ,

[0135] Wherein, ΔT adjust (t + k) is the adjustment range of the battery temperature change, indicating the difference between the predicted future temperature and the current actual temperature, and T current is the actual temperature of the current battery;

[0136] If Δ Tadjust (t + k)>0, it indicates that the future temperature is too high, and the cooling device is started; if Δ Tadjust (t + k)<0, it indicates that the temperature is too low, and the heating device is started. Further, the control system sets the power of the temperature control device according to the temperature adjustment range ΔT adjust to ensure that the temperature change is within the predetermined range. The specific temperature control strategy is adjusted through the following model:

[0137] P cooling = α × ΔT adjust (t + k),

[0139] P heating = β × |ΔT adjust (t + k)|,

[0141] Wherein, α and β are the adjustment coefficients of the cooling and heating systems, which are set according to the response ability of the device and the target temperature range, and P cooling is the power required to start the cooling system, and P heating is the power required to start the heating system.

[0142] In this way, the control system can pre-regulate the battery temperature in advance to ensure that the battery operates at the optimal working temperature.

[0143] Specific Embodiment 1: In this embodiment, the battery monitoring device accurately measures the voltage, current, and temperature of the battery pack in the ship container, collects and transmits the real-time data to the central control system. These data are the basis for ensuring the stable and safe operation of the battery pack. Real-time monitoring of the battery pack status can promptly detect abnormal battery temperatures and avoid failures caused by overheating or overcooling. The voltage and current data of the battery can reflect the charging and discharging conditions of the battery, further helping to analyze the load level and energy efficiency of the battery under different operating modes, while the temperature data is the most critical parameter, directly affecting the chemical reaction rate, battery life, and safety of the battery.

[0144] When the battery temperature data collected by the battery monitoring device exceeds the set safety threshold, the central control system will activate the cooling system for temperature reduction. The cooling system can be a liquid cooling system or an air cooling system, and the specific choice depends on the design of the ship system and the working environment of the battery pack. The liquid cooling system uses a liquid medium (such as water or oil) to flow through the heat dissipation device of the battery pack to carry away the heat, thereby reducing the temperature of the battery pack; while the air cooling system enhances air flow through a powerful fan or ventilation system to quickly dissipate heat. During high load, high temperature or rapid charge and discharge processes, the cooling system can effectively prevent the battery from overheating and avoid battery aging, reduced efficiency or safety accidents caused by excessive battery temperature.

[0145] On the contrary, when the battery temperature is lower than the set minimum safety threshold, the system will activate the heating system for temperature increase. The function of the heating system is to ensure that the battery can still maintain an ideal working temperature in a low-temperature environment, thereby avoiding problems such as a decrease in battery capacity and a reduction in discharge efficiency caused by low temperature. The heating device can adopt methods such as an electric heater, a hot air system or a heating film on the battery surface to provide a heat source around the battery to ensure that the temperature of the battery pack remains within an appropriate range.

[0146] During this process, the temperature regulation system not only needs to respond to the changes in battery temperature in real time, but also needs to make corresponding adjustments according to the charge and discharge status of the battery pack, external environmental conditions and the navigation mode of the ship. The system dynamically monitors temperature, load and environmental changes and adjusts the operating modes of the cooling and heating devices in a timely manner. For example, when the ship sails to a high-temperature area, the response time of the cooling system will be shorter and the cooling intensity will be higher; while when sailing in a cold area, the system will activate the heating device in advance to ensure that the battery temperature does not drop too low.

[0147] In addition, the control system will also dynamically optimize the temperature regulation through an accurate feedback mechanism. During actual operation, the battery monitoring device will report the changes in battery temperature to the system in real time, and the control system will continuously adjust the working mode of the temperature control device according to the data feedback. The start and stop of the cooling or heating device should be precise to avoid unnecessary energy waste caused by over-regulation and improve the overall energy efficiency of the system. This strategy combining temperature control and energy efficiency optimization can greatly reduce the energy consumption of the battery management system, improve the working efficiency of the battery, thereby reducing the operating cost of the ship and extending the service life of the battery pack.

[0148] During long-term use, the system can also continuously optimize the temperature control strategy based on historical data. The control system can self-learn and improve the temperature control strategy by analyzing the battery temperature change trend during different navigation stages of the ship and the operating effect of the temperature control device. For example, when the ship operates in a specific area, the system can automatically adjust the working parameters of the cooling or heating system to adapt to the new environmental temperature change, thereby achieving more intelligent battery management.

[0149] Specific Embodiment 2: In this embodiment, a machine learning model is adopted to enhance the intelligence level of the temperature control system, enabling the battery management system to predict future changes in battery temperature and start the temperature control device in advance, thereby achieving more precise and efficient temperature regulation. First, the battery monitoring device and the environmental sensor will collect battery voltage, current, temperature, and external environmental data (such as temperature, humidity, etc.) in real time. These data will be transmitted to the central control system for real-time analysis and training of the machine learning model.

[0150] The machine learning model is trained based on a large amount of historical data and can identify the complex relationships between battery temperature, environmental changes, and charge / discharge states. For example, by learning the battery temperature changes under different environmental temperatures, humidities, and load conditions, the machine learning model can predict the future temperature change trend of the battery within a certain period. When the battery temperature approaches the set threshold, the system will start the cooling or heating device in advance to prevent the battery temperature from exceeding the safe range. This method is more precise than the traditional threshold-based control system because it not only relies on current data but also comprehensively considers historical data and future trends, thus making temperature control adjustments in advance.

[0151] For example, during high-load charging or discharging, the temperature of the battery may rise rapidly. The traditional temperature control system may only start the cooling system after the temperature exceeds the threshold, while the intelligent system based on machine learning can predict the future temperature change trend according to multiple factors such as the current charging speed, battery health status, and external environmental temperature, and start the cooling device in advance. This predictive control can avoid overheating and reduce the response time of the temperature control system, improving battery safety.

[0152] The machine learning model will continuously learn new data during long-term operation and gradually improve the prediction accuracy. By deeply analyzing the battery pack, environmental changes, and operating modes, the system can adjust its temperature control strategy to adapt to different ship navigation conditions and battery usage scenarios. This self-optimizing ability cannot be achieved by the traditional threshold-based control system.

[0153] In addition, the intelligent temperature control system based on machine learning can monitor the health status of the battery in real time and make corresponding temperature control adjustments according to the aging state of the battery. For example, as the usage time of the battery pack increases, the thermal stability of the battery may decline, and the system can adjust the cooling or heating strategy based on this information to ensure that the battery can maintain a stable operating temperature even during the aging process, avoiding performance degradation caused by abnormal temperature.

[0154] Specific implementation 3: This implementation proposes a temperature control scheduling method based on comprehensive energy efficiency optimization, which combines battery temperature regulation with overall ship energy efficiency optimization to achieve dual optimization of temperature control and energy saving. The battery monitoring device obtains the voltage, current and temperature data of the battery pack in real time, and combines the environmental data collected by the external environmental sensor. The system will analyze the current operating status of the battery and the ship based on these data and dynamically optimize the temperature control strategy. In this way, the system not only considers the temperature data of the battery, but also decides when to start the cooling or heating equipment and how to adjust the working intensity of these equipment according to the battery charging status, load conditions and external environmental factors to achieve optimal energy efficiency.

[0155] When the battery load is light or the battery power is high, the system will choose to reduce the cooling or heating power to reduce energy consumption; when the battery load is heavy or the battery power is low, the system will appropriately increase the cooling or heating power to ensure that the battery is in the optimal operating temperature range. At this time, the temperature control scheduling system can not only ensure that the battery temperature remains within a safe range, but also optimize the use of energy according to the battery status and the ship's navigation conditions, maximizing the battery's working efficiency.

[0156] This comprehensive energy efficiency optimized temperature control scheduling system can make intelligent adjustments according to the ship's navigation route and the scheduled navigation mode. For example, when the ship enters a high temperature area, the workload of the cooling system will automatically increase to prevent the battery temperature from rising; in a low temperature environment, the system will start the heating equipment in advance to prevent the battery from overcooling. This intelligent scheduling method not only ensures the safe operation of the battery, but also dynamically optimizes the temperature control strategy according to the real-time navigation status to improve the overall energy efficiency of the ship.

[0157] In addition, the system continuously optimizes the temperature control scheduling strategy through the accumulation of historical data. During long-term operation, the control system will analyze the laws of battery temperature changes and the influence of various environmental factors, and then optimize the response of the temperature control system. In this way, it can not only improve the stability of the battery management system, but also improve the overall operating efficiency of the ship and realize intelligent energy management. This temperature control scheduling solution based on comprehensive energy efficiency optimization makes the ship's energy management more efficient and accurate, thereby reducing the ship's operating costs and maximizing the service life of the battery pack.

[0158] Through real-time monitoring of the temperature, charge and discharge status, and external environmental conditions of the battery pack, and combined with the precise adjustment of the cooling and heating systems, the present invention can significantly improve the safety and stability of the battery. Traditional battery management systems mainly rely on fixed temperature thresholds for control, but this method may suffer from response lags or insufficient temperature control. By adopting a dynamic temperature control scheme (such as a temperature control scheduling system based on battery monitoring devices, machine learning models, or comprehensive energy efficiency optimization), the system can judge the working state of the battery in real time and make precise temperature adjustments according to the actual needs of the battery and changes in the external environment. Timely activate the cooling or heating equipment to avoid extreme situations of overheating or overcooling of the battery, thereby effectively preventing performance degradation, capacity decline, and even safety accidents caused by excessive temperature fluctuations of the battery.

[0159] At the same time, the temperature control system can adaptively adjust according to the charge and discharge load of the battery pack, environmental conditions, and the health status of the battery, reduce the amplitude of temperature fluctuations, slow down the battery aging process, and extend the service life of the battery. By optimizing temperature management, the chemical reactions of the battery and the working environment of the internal components of the battery are better protected, avoiding the acceleration of chemical reactions caused by high temperatures or the decrease in battery efficiency caused by low temperatures. This intelligent temperature control strategy not only extends the battery life cycle but also greatly reduces battery failures and maintenance costs caused by improper temperature management.

[0160] Based on the combination of a temperature control scheduling system and comprehensive energy efficiency optimization, the present invention enables the ship to utilize energy more efficiently during operation, maximize the energy efficiency of the battery, and thereby reduce the overall operating cost. During the ship's navigation, the system dynamically adjusts the working intensity of the temperature control system according to the battery load, charge and discharge status, and external environmental changes to avoid excessive energy consumption. This not only reduces the energy waste of the cooling or heating equipment but also avoids the energy efficiency loss caused by too high or too low battery temperature. For example, when the battery load is light, the system will reduce the power of cooling or heating to save energy; while in high-load or low-battery situations, the system will appropriately increase the power of the temperature control equipment to ensure that the battery operates in the best working state.

[0161] In addition, the introduction of machine learning and optimization scheduling algorithms makes the temperature control system more intelligent and precise, capable of predicting battery temperature changes in advance and taking preventive measures to reduce energy waste. Through the analysis of historical data and real-time feedback, the system can gradually optimize the temperature control strategy, improve energy utilization efficiency, and reduce unnecessary energy consumption. In the long run, this intelligent energy management not only improves the overall operating efficiency of the ship but also effectively reduces energy costs, reduces the negative impact on the environment, and promotes the green and sustainable operation of the ship.

[0162] Only certain exemplary embodiments of the present invention have been described by way of illustration above. Without doubt, for those of ordinary skill in the art, the described embodiments can be modified in various different ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An energy-saving control method for an electric ship container, characterized in that: The following steps are involved: By installing a battery monitoring device, the voltage, current and temperature of the battery pack in the ship container are monitored in real time, and data is collected at a preset frequency to form a battery status data set; Through environmental sensors, the external environment temperature and humidity during the ship's operation are collected in real time to generate environmental data sets; Based on the temperature data in the battery status data set, when the battery temperature exceeds the preset safety threshold, the cooling system is started to cool down the battery; when the battery temperature is lower than the preset threshold, the heating system is started to heat up the battery to keep the battery within the ideal operating temperature range; Based on the optimization scheduling algorithm that minimizes energy consumption, the battery pack's charge and discharge strategy is dynamically adjusted according to the battery pack's real-time power, charge and discharge status, and environmental data to achieve optimal energy efficiency; Based on the trained machine learning model, the future change trend of battery temperature is predicted, and the operation strategy of the cooling or heating equipment is adjusted in advance to ensure that the battery is always within a safe temperature range.

2. The energy-saving control method for an electric ship container according to claim 1 is characterized in that: The specific steps to form a battery status data set by installing a battery monitoring device to monitor the voltage, current and temperature of the battery pack in the ship container in real time and collect data at a preset frequency are as follows: First, the battery monitoring device is connected to the electrical interface of the battery pack to obtain the voltage, current and temperature data of the battery pack in real time; The battery monitoring device processes the battery voltage, current and temperature data collected in real time according to the preset collection frequency; The data transmitted to the control system in real time will be processed by the data processing module to form a battery status data set with timeliness; The processed battery status data will be stored in a central control system or cloud server to form a battery health monitoring database.

3. The energy-saving control method for an electric ship container according to claim 1 is characterized in that: The environmental sensors are used to collect the external environment temperature and humidity in real time during the operation of the ship. The specific steps to generate the environmental data set are as follows: While the battery monitoring device monitors the battery status in real time, the environmental sensor begins to collect real-time data on the external environment temperature and humidity during the operation of the ship; The temperature and humidity data collected by the environmental sensors will be collected at a preset frequency and transmitted to the central control system; Once the environmental data is collected and transmitted to the control system, it will be processed and analyzed together with the battery status data; The processed environmental data will be stored in the database of the central control system to form a complete environmental data set.

4. The energy-saving control method for an electric ship container according to claim 1, characterized in that: Based on the temperature data in the battery status data set, when the battery temperature exceeds the preset safety threshold, the cooling system is started for cooling; when the battery temperature is lower than the preset threshold, the heating system is started for heating. The specific steps to keep the battery within the ideal operating temperature range are as follows: The central control system compares the temperature data in the battery status data set with the preset safety threshold in real time; Once it is determined that the battery temperature exceeds the safe range, the control system will automatically adjust the working state of the cooling or heating equipment according to the temperature change; After the cooling or heating system is started, the central control system will continue to monitor changes in battery temperature and adjust the temperature control strategy in real time.

5. The energy-saving control method for an electric ship container according to claim 1 is characterized in that: Based on the optimization scheduling algorithm that minimizes energy consumption, the specific steps for dynamically adjusting the charging and discharging strategy of the battery pack to achieve the optimal energy utilization efficiency are as follows: First, we need to obtain the real-time power and charge and discharge status of the battery pack. These data are the basis for optimizing the charge and discharge strategy. The calculation expression is as follows: ， In the formula, E real is the actual current capacity of the battery pack, I battery (t) is the instantaneous current of the battery, i.e., the instantaneous current at time point t; Charge and discharge state C status By monitoring the direction and strength of the battery current, if the current is positive, the battery is discharging, and if the current is negative, the battery is charging; When formulating the charging and discharging strategy, environmental data has a direct impact on the energy efficiency and performance of the battery. Therefore, environmental data is taken into consideration. According to the impact of temperature and humidity on battery performance, a correction factor is set to adjust the battery charging and discharging efficiency. The calculation expression is as follows: f temp =1+α temp (T env -T opt ), f hum =1+β hum (H env -H opt ), In the formula, f temp is the ambient temperature correction factor, T env is the external ambient temperature, T opt is the optimal operating temperature of the battery, α temp is the influence coefficient of temperature on battery charging and discharging efficiency, f hum is the ambient humidity correction factor, H env is the external environment humidity, H opt is the optimal operating humidity of the battery, β hum is the influence coefficient of humidity on battery charging and discharging efficiency, Based on the actual current charge E of the battery pack real , charge and discharge state C status , and the ambient temperature correction factor f temp and the ambient humidity correction factor f hum , dynamically adjust the charging and discharging power distribution by optimizing the scheduling algorithm, and the calculation expression is as follows: ， Where, L energy is the energy loss of the battery, E battert is the total charge of the battery, P charging (t) is the charging power of the battery at time t; To further improve energy efficiency, the optimization scheduling algorithm also needs to adjust the charging and discharging strategy by predicting the future trend of battery power. The battery power trend is modeled using a machine learning model or a time series prediction method to predict the battery power change in the future. The calculation expression is as follows: ， In the formula, E predict (t+Δt) is the predicted battery power, that is, the predicted value of the battery power within the time interval Δt, E real (t) is the actual battery charge at time t, P net (t) is the battery network power, the net power of the battery at time point t.

6. The energy-saving control method for an electric ship container according to claim 1, characterized in that: Based on the trained machine learning model, the future trend of battery temperature is predicted, and the operation strategy of the cooling or heating equipment is adjusted in advance to ensure that the battery is always within a safe temperature range. The specific steps are as follows: In order to train the machine learning model and perform temperature prediction, it is necessary to first collect a large amount of historical battery temperature data. This data will be used as input features. The real-time data collected by the battery monitoring device and environmental sensors will form a data set and be input into the machine learning model. To ensure the accuracy and availability of the data, the data needs to be preprocessed before training. Suppose the collected battery temperature data is T(t), the ambient temperature data is E(t), and the battery voltage and current data are V(t) and I(t), respectively, where t represents the time point; The data set after data preprocessing will form the basis of model training, ensuring that the model can learn the law of temperature change more accurately. For each time point t, the historical data of temperature change is constructed into a feature set X(t) = [T(tk), E(tk)]k), V(tk), I(tk)], where k is the size of the sliding window, indicating the range of historical data used for prediction, T(tk) is the battery temperature history data, E(tk) is the ambient temperature history data, V(tk) is the battery voltage history data, and I(tk) is the battery current history data; Use the preprocessed data set to train a machine learning model. By fitting the relationship between battery temperature and its related features, the future trend of battery temperature is predicted. During the training process, the parameters of the model are optimized by minimizing the loss function. The expression is as follows: ， Where L is the minimization loss function, T(t+k) is the actual battery temperature, is the battery temperature predicted by the model, is the battery temperature at the future time (t+k) predicted by the machine learning model based on the current input features, and N is the number of samples; The machine learning model is trained and used to predict the battery temperature trend at the future time point t+k. By inputting the current state feature X(t), the model will output the predicted value of the future temperature. (t+k), suppose the temperature needs to be predicted N steps in the future, the model will output the predicted temperature for each time step in turn Model Output It is a temperature prediction based on the state characteristics at the current moment. The formula is as follows: ， Where f(X(t)) is the model function, which represents the machine learning model and maps the input feature set X(t) to the battery temperature prediction value superior; According to the temperature prediction results of the machine learning model, the operation strategy of the cooling or heating equipment is adjusted in advance. When it is predicted that the battery temperature will exceed the upper threshold T in the future, max When the predicted temperature is lower than the lower threshold T min When the heating device is started, in order to ensure that the battery temperature is always kept within a safe range, the control system makes dynamic adjustments based on the prediction results, using the following formula: ， In the formula, ΔT adjust (t+k) is the adjustment amplitude of the battery temperature change, which represents the difference between the predicted future temperature and the current actual temperature. current is the actual temperature of the current battery; If ΔT adjust (t+k)>0, indicating that the temperature is too high, start the cooling equipment; if ΔT adjust (t+k)<0, indicating that the temperature is too low, the heating equipment is started, and further, the control system adjusts the amplitude ΔT according to the temperature. adjust Set the power of the temperature control device to ensure that the temperature change is within the predetermined range. The specific temperature control strategy is adjusted through the following model: P cooling =α×ΔT adjust (t+k), P heating =β×|ΔT adjust (t+k)|, Where α and β are the adjustment coefficients of the cooling and heating systems, P cooling is the power required to start the cooling system, P heating is the power required to start the heating system.

Images