Vehicle transportation route hydrogen consumption simulation system and method based on multiple models

Abstract

The invention provides a vehicle transportation route hydrogen consumption simulation system and method based on multiple models. The method belongs to the technical field of new energy and simulation, and comprises the steps of based on a basic power demand model and a hydrogen consumption calculation model, collecting multi-source data in real time through a multi-source data collection network, combining the multi-source data with historical data, optimizing an existing model through a machine learning algorithm, and updating the hydrogen consumption calculation model. By collecting multi-source data in real time and optimizing an existing model in combination with historical data and a machine learning algorithm, the hydrogen consumption condition can be predicted more accurately. Meanwhile, by collecting new data of external environmental factors in real time, the environmental influence model is optimized, the hydrogen consumption calculation result is further corrected, and the simulation precision is improved.

Description

Technical Field

[0001] The invention proposes a vehicle transportation route hydrogen consumption simulation system and method based on multiple models, belonging to the technical field of new energy and simulation. Background Art

[0002] As a clean and efficient form of energy, hydrogen energy is gaining increasing attention in the field of transportation. Especially in vehicle transportation, hydrogen fuel cell vehicles are regarded as an important direction for future transportation development due to their advantages such as zero emissions and long driving range. However, the hydrogen consumption of hydrogen fuel cell vehicles directly affects their operating costs and environmental benefits. Therefore, accurately predicting and optimizing the hydrogen consumption of vehicles on different transportation routes has become an urgent problem to be solved.

[0003] Traditional hydrogen consumption prediction methods mainly rely on theoretical calculations and experimental tests. Although these methods can reflect the hydrogen consumption of vehicles to a certain extent, they often have problems such as insufficient accuracy and poor adaptability. Especially in complex actual operating environments, the vehicle's power demand, the efficiency of hydrogen fuel cells, and external environmental factors will have a significant impact on hydrogen consumption. The interaction between these factors makes hydrogen consumption prediction more complicated.

[0004] In order to overcome the limitations of traditional methods, in recent years, multi-model-based simulation methods have gradually been introduced into the prediction of hydrogen consumption for vehicle transportation routes. This method integrates multiple related models and comprehensively considers multiple factors such as vehicle power requirements, hydrogen fuel cell efficiency, and external environmental factors to achieve accurate prediction of hydrogen consumption. However, the existing multi-model simulation method still has some shortcomings, such as untimely model updates, poor adaptability to complex environments, and large deviations between simulation results and actual operating data. Summary of the invention

[0005] The present invention provides a vehicle transportation route hydrogen consumption simulation system and method based on multiple models to solve the problems mentioned in the above background technology:

[0006] The present invention proposes a multi-model based vehicle transportation route hydrogen consumption simulation method, the method comprising:

[0007] S1. Based on the existing basic power demand model and hydrogen consumption calculation model, multi-source data collected in real time through a multi-source data acquisition network is combined with historical data, the existing model is optimized through a machine learning algorithm, and the hydrogen consumption calculation model is updated;

[0008] S2. Optimize the existing environmental impact model through new data of external environmental factors collected in real time, and further correct the hydrogen consumption calculation results according to the impact of environmental conditions on hydrogen fuel cell efficiency;

[0009] S3. Integrate the optimized basic power demand model, hydrogen consumption calculation model and environmental impact model to form a multi-level comprehensive hydrogen consumption model;

[0010] S4. Input relevant data of different transportation routes, perform simulation calculations on different transportation routes based on the comprehensive hydrogen consumption model, and obtain the hydrogen consumption of each route under different working conditions; output the simulation results of hydrogen consumption of each transportation route under different working conditions, and continuously optimize the model by comparing the actual operation data;

[0011] S5. Based on the simulation results, the optimal transportation route is determined through a multi-objective optimization algorithm; the optimized optimal transportation route suggestion is output, and the transportation route is dynamically adjusted according to the actual operation situation;

[0012] S6. During the transportation process, actual operation data is collected in real time; based on the actual operation data, the comprehensive hydrogen consumption model is re-evaluated and optimized.

[0013] The present invention proposes a vehicle transportation route hydrogen consumption simulation system based on multiple models, the system comprising:

[0014] Data acquisition module: Based on the existing basic power demand model and hydrogen consumption calculation model, multi-source data is collected in real time through a multi-source data acquisition network; the multi-source data is combined with historical data, the existing model is optimized through a machine learning algorithm, and the hydrogen consumption calculation model is updated;

[0015] Model building module: optimizes the existing environmental impact model through new data of external environmental factors collected in real time, and further corrects the hydrogen consumption calculation results according to the impact of environmental conditions on hydrogen fuel cell efficiency;

[0016] Model integration module: Integrate the optimized basic power demand model, hydrogen consumption calculation model and environmental impact model to form a multi-level comprehensive hydrogen consumption model;

[0017] Result simulation module: input relevant data of different transportation routes, simulate and calculate different transportation routes based on the comprehensive hydrogen consumption model, and obtain the hydrogen consumption of each route under different working conditions; output the simulation results of hydrogen consumption of each transportation route under different working conditions, and continuously optimize the model by comparing with the actual operation data;

[0018] Route optimization module: Based on the latest simulation results, the optimal transportation route is determined through a multi-objective optimization algorithm; the optimal transportation route recommendation is output, and the transportation route is dynamically adjusted according to the actual operation situation;

[0019] Dynamic correction module: During the transportation process, actual operation data is collected in real time; based on the actual operation data, the comprehensive hydrogen consumption model is re-evaluated and optimized.

[0020] Beneficial effects of the present invention: by collecting multi-source data in real time, optimizing the existing model by combining historical data and machine learning algorithms, the hydrogen consumption can be predicted more accurately. At the same time, by collecting new data of external environmental factors in real time, the environmental impact model is optimized, the hydrogen consumption calculation results are further corrected, and the simulation accuracy is improved; according to the actual operation data and the comprehensive hydrogen consumption model, simulation calculations are performed on different transportation routes to obtain the hydrogen consumption of each route under different working conditions. By continuously optimizing the model by comparing the actual operation data, the transportation route can be dynamically adjusted according to the actual situation and the best transportation route can be selected; the best transportation route is determined by a multi-objective optimization algorithm, the optimized best transportation route recommendation is output, and the transportation route is dynamically adjusted according to the actual operation situation. This helps to improve transportation efficiency and reduce transportation costs; during the transportation process, the actual operation data is collected in real time, and the comprehensive hydrogen consumption model is re-evaluated and optimized according to the actual operation data. This helps to find problems in a timely manner and make adjustments to ensure the safety and efficiency of the transportation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a step diagram of the method of the present invention;

[0022] Figure 2 This is a system module diagram of the present invention. DETAILED DESCRIPTION

[0023] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0026] One embodiment of the present invention, as Figure 1 As shown, a method for simulating hydrogen consumption of a vehicle transportation route based on multiple models, the method comprising:

[0027] S1. Based on the existing basic power demand model and hydrogen consumption calculation model, multi-source data is collected in real time through a multi-source data collection network; the multi-source data is combined with historical data, the existing model is optimized through a machine learning algorithm, and the hydrogen consumption calculation model is updated;

[0028] S2. Optimize the existing environmental impact model through new data of external environmental factors collected in real time, and further correct the hydrogen consumption calculation results according to the impact of environmental conditions on hydrogen fuel cell efficiency;

[0029] S3. Integrate the optimized basic power demand model, hydrogen consumption calculation model and environmental impact model to form a multi-level comprehensive hydrogen consumption model;

[0030] S4. Input relevant data of different transportation routes, perform simulation calculations on different transportation routes based on the comprehensive hydrogen consumption model, and obtain the hydrogen consumption of each route under different working conditions; output the simulation results of hydrogen consumption of each transportation route under different working conditions, and continuously optimize the model by comparing the actual operation data;

[0031] S5. Based on the simulation results, the optimal transportation route is determined through a multi-objective optimization algorithm; the optimized optimal transportation route suggestion is output, and the transportation route is dynamically adjusted according to the actual operation situation;

[0032] S6. During the transportation process, actual operation data is collected in real time; based on the actual operation data, the comprehensive hydrogen consumption model is re-evaluated and optimized.

[0033] The working principle of the above technical solution is: through sensors installed on the vehicle and other data sources (such as GPS, weather stations, etc.), the vehicle's operating parameters such as speed, acceleration, slope, and data on external environmental factors such as temperature, humidity, and altitude are collected in real time. These new data are combined with historical data and trained through machine learning algorithms to optimize the existing basic power demand model and hydrogen consumption calculation model. Among them, the multi-source data acquisition network refers to a network system that obtains information from multiple different data sources. For example, sensors inside the vehicle can provide vehicle status data, while public meteorological services or dedicated environmental monitoring stations can provide external environmental conditions. New data on external environmental factors collected in real time are used to further optimize the environmental impact model. This model takes into account how environmental changes affect the efficiency of hydrogen fuel cells and adjusts the hydrogen consumption calculation results according to the latest environmental conditions. This ensures that the hydrogen consumption forecast is closer to the actual situation. Among them, the environmental impact model is a mathematical or physical model that describes how external environmental factors (such as temperature and humidity) affect the efficiency of hydrogen fuel cells. For example, low temperatures may reduce the reaction rate of fuel cells, thereby increasing hydrogen consumption. The optimized basic power demand model and hydrogen consumption calculation model in S1 are combined with the optimized environmental impact model in S2 to form a multi-level comprehensive hydrogen consumption model. This model can more accurately simulate the hydrogen consumption of vehicles under different conditions. Input relevant data of various transportation routes, such as road type, traffic flow, weather forecast, etc., and use the optimized comprehensive hydrogen consumption model to simulate and calculate the hydrogen consumption of different routes. The simulation results not only help plan the most economical transportation route, but also provide an opportunity to continuously optimize the model by adjusting the model parameters by comparing the differences between actual operation data and simulation results. Based on the simulation results in S4, a multi-objective optimization algorithm is used to determine the best transportation route. The multi-objective optimization algorithm takes into account multiple objectives (such as minimizing hydrogen consumption, shortening transportation time, etc.) and finds a balance point, namely the Pareto optimal solution. The final output is a recommendation for the best transportation route that can save fuel and meet other operational requirements. Throughout the transportation process, actual operation data continues to be collected in real time, including the vehicle's location, speed, hydrogen consumption, etc. The comprehensive hydrogen consumption model is then re-evaluated and optimized based on these actual data to ensure that it always reflects the latest reality. This process is cyclical, and as new data is added, the model becomes more and more accurate.

[0034] The effect of the above technical solution is: by combining multi-source data collected in real time with historical data and using machine learning algorithms to optimize existing models, hydrogen consumption under different working conditions can be more accurately predicted. This helps logistics companies better plan and manage their operating costs; the continuous optimization of the environmental impact model takes into account the impact of external conditions such as temperature, humidity, and altitude on the efficiency of hydrogen fuel cells, ensuring that reliable hydrogen consumption estimates can be provided under various climatic and geographical conditions, enhancing the flexibility and scope of application of the system; the multi-level comprehensive hydrogen consumption model integrates multiple aspects such as basic power demand, hydrogen consumption calculation, and environmental impact, providing a more comprehensive and detailed analysis tool to support logistics companies in fine management and decision-making in daily operations; based on the latest simulation results and multi-objective optimization algorithms, it can provide logistics companies with the best transportation route recommendations. This method not only takes into account the minimization of hydrogen consumption, but also takes into account other key factors (such as time and safety) at the same time, thereby improving overall operational efficiency; in the actual transportation process, the vehicle operation status information is continuously collected through the real-time data acquisition system, and the comprehensive hydrogen consumption model is re-evaluated and optimized accordingly. This feature enables the model to respond quickly to changing conditions and maintain long-term effectiveness and accuracy; more accurate hydrogen consumption estimates and the selection of optimal transportation routes directly lead to lower fuel consumption and operating costs, reducing unnecessary expenses and increasing the economic benefits of the enterprise; hydrogen fuel cell vehicles themselves are a clean energy transportation tool, and this simulation method further promotes the application and development of this technology, which is conducive to reducing greenhouse gas emissions and other pollutants and promoting the development of green logistics; through scientific planning and dynamic adjustment of transportation routes, transportation time and uncertainty can be reduced, and the on-time delivery rate of goods can be improved.

[0035] In one embodiment of the present invention, the S1 includes:

[0036] S11. Define the type and location of data collection points, and design data collection frequency and accuracy requirements;

[0037] S12. Build a data transmission network, encrypt the data transmission process based on the data encryption algorithm, integrate the data collection system, and uniformly receive, store and manage multi-source data;

[0038] S13, obtaining historical data, and preprocessing the collected real-time data and historical data;

[0039] S14. Based on the existing vehicle motion equation model and vehicle dynamics principles, combined with newly collected real-time vehicle operation data, optimize and calculate the basic power requirements under different working conditions;

[0040] S15. Based on the existing power system response characteristic curve and efficiency curve model, the new hydrogen fuel cell working characteristic data is used to optimize the relationship between hydrogen consumption and electric energy output, and an updated hydrogen consumption calculation model is established;

[0041] S16. Combine the existing basic power demand model and hydrogen consumption calculation model, and further optimize it through new data to more accurately calculate the hydrogen consumption under different working conditions. The hydrogen consumption, that is, the hydrogen consumption rate, is calculated by the following formula:

[0042]

[0043] in, Indicates the power demand of the vehicle under a certain working condition. represents the overall efficiency of the hydrogen fuel cell system, Indicates the operating voltage of the hydrogen fuel cell, represents the Faraday constant; Indicates the start time of hydrogen consumption, Indicates the end time of hydrogen consumption.

[0044] The working principle of the above technical solution is to determine which types of sensors will be installed where, as well as the frequency and accuracy of data collection of these sensors. For example, information such as speed, acceleration, slope, etc. can be obtained through on-board sensors, while external environmental factors such as temperature and humidity may be collected through weather stations or other environmental monitoring equipment near the vehicle. The data collection point refers to the specific location where the sensor is installed. For example, a speed sensor is installed on the wheel, a temperature and humidity sensor is installed in the car, and a slope sensor is installed on the vehicle chassis. The data collection frequency refers to how many times data is collected per second or per minute. For example, for key parameters such as speed, a higher collection frequency (such as 10 times per second) may be required, while for slower changing factors such as temperature, a lower frequency (such as once per minute) may be required; the accuracy requirement refers to the demand for the accuracy of the measurement results. For example, a slope sensor may need to achieve an accuracy of ±0.5 degrees to ensure that it can accurately reflect the degree of inclination of the road. Establish a secure and reliable data transmission network and use encryption algorithms to protect data transmission from the vehicle to the data center. At the same time, integrate a data collection system to receive data from different sources and centrally store and manage them. The data transmission network can be a cellular network, Wi-Fi, or a dedicated wireless network, etc., to ensure that data can be quickly and stably transmitted from the vehicle to the server. Data encryption algorithms such as AES (Advanced Encryption Standard) and RSA are used to ensure the security of data in the network and prevent unauthorized access. Collect past operation records as historical data, and then clean, format, and perform other necessary preprocessing operations on the newly acquired real-time data and these historical data so that they can be used to train machine learning models or directly participate in calculations. Use existing physical models to describe the motion characteristics of the vehicle, and adjust these models based on the latest real-time data to more accurately predict the energy required under various driving conditions. For example, when climbing a slope, the vehicle requires additional power support, which can be more accurately estimated by the updated model. The vehicle motion equation model usually contains formulas including Newton's second law to express the power requirements of the vehicle when accelerating, decelerating, turning, etc. For example, F=ma (force equals mass multiplied by acceleration). The operating condition refers to the operating state of the vehicle under specific conditions, such as driving at a constant speed on a flat road, accelerating to overtake, decelerating to stop, climbing a slope, etc. Using the established power system performance model, combined with the latest working characteristics of hydrogen fuel cells (such as temperature, pressure, current density, etc.), the relationship between hydrogen consumption and power output is re-evaluated and optimized to obtain a more accurate hydrogen consumption calculation model. Among them, the response characteristic curve shows how the power system responds to different input conditions over time, such as the reaction speed of the system when the load increases. The efficiency curve model describes the proportion of the input energy converted into useful work by the system under different operating conditions. For example, as the temperature increases, the efficiency of the hydrogen fuel cell may decrease, resulting in more hydrogen consumption.The optimized basic power demand model and hydrogen consumption calculation model are integrated to form a comprehensive framework that can dynamically adjust its parameters based on the latest vehicle operation data to achieve high-precision prediction of hydrogen consumption under different driving conditions.

[0045] The effects of the above technical solutions are as follows: by clarifying the type and installation location of the sensor, it is ensured that the collected data can fully reflect the vehicle's operating status and environmental conditions; by reasonably setting the data collection frequency and accuracy, it can not only ensure the validity of the data, but also avoid unnecessary data redundancy and computing burden; high-precision data collection helps to improve the accuracy of subsequent analysis, thereby improving the reliability and stability of the entire system; using encryption algorithms to protect the security of data transmission, prevent sensitive information from being leaked or tampered with, and protect the business secrets of logistics companies; integrating data from different sources onto one platform simplifies the data management and analysis process and improves work efficiency; establishing a stable data transmission network, supporting real-time data upload and remote monitoring, and facilitating rapid response to abnormal situations; combining historical data and real-time data to provide a more complete analysis perspective, helping to discover long-term trends and short-term fluctuations; through pre-processing steps such as cleaning and formatting, noise and error values ​​are eliminated to improve the quality and availability of data; standardized data can be directly used for modeling and simulation, reducing data preparation time and speeding up decision-making; updated models can more accurately predict the vehicle's performance in various working conditions (such as climbing, accelerating The power demand under the condition of high speed (speed) can be used to better plan fuel use; the power demand model can be customized according to the specific characteristics of different models to make the simulation results closer to the actual situation; the optimized model helps to identify the most economical driving mode, reduce energy consumption and improve operational efficiency; when the vehicle enters an uphill section, the model will adjust the power demand prediction according to the latest slope sensor data to ensure that there is enough power to support climbing while minimizing hydrogen consumption; the model can automatically adjust according to the working characteristics of hydrogen fuel cells and changes in environmental factors to maintain a high prediction accuracy; through the refined management of the relationship between hydrogen consumption and power output, the optimal energy distribution is achieved and unnecessary fuel waste is reduced; the continuously updated model can timely reflect the impact of factors such as equipment aging on efficiency and help maintain optimal performance; by continuously introducing new data, the model can self-learn and optimize, always keep the latest status and adapt to the changing operating environment; the comprehensive model can more accurately simulate the hydrogen consumption under different working conditions, providing a reliable basis for the selection of transportation routes; the model has adaptive capabilities and can be adjusted in real time according to actual operating conditions to ensure that each prediction is based on the latest operating conditions. The above formula combines the principles of vehicle dynamics with real-time operating data. The formula can accurately calculate the power requirements under different operating conditions, including factors such as vehicle weight, wind resistance, rolling resistance, and climbing resistance, ensuring the accuracy of the calculation results. Real-time data collection allows the formula to dynamically adjust the calculation to adapt to changes in vehicle operation, such as speed, acceleration, slope, etc. Combining the response characteristics and efficiency curve of the hydrogen fuel cell system, the power demand can be corrected to optimize the hydrogen consumption rate and improve energy efficiency. By accurately calculating hydrogen consumption, the operation of the vehicle power system can be optimized and unnecessary energy consumption can be reduced.Combining the basic power demand model with the hydrogen consumption calculation model provides a unified calculation framework, which improves the reliability and stability of the overall system. By preprocessing historical data, the formula can learn and adapt to the performance of the vehicle under different conditions, enhancing the reliability of the model. By reducing the waste of hydrogen fuel, operating costs can be reduced, which means lower operating costs and higher economic benefits for vehicle operators. Accurate hydrogen consumption calculation helps to optimize maintenance plans, reduce unnecessary maintenance operations, and reduce maintenance costs. By improving the efficiency of hydrogen fuel cells and reducing hydrogen consumption, possible environmental pollution and greenhouse gas emissions are indirectly reduced. Accurate calculation of hydrogen consumption helps to optimize the use of hydrogen energy and promote the rational use of resources. Through accurate calculation of power demand, overload operation of hydrogen fuel cell systems can be avoided, system life can be extended, and vehicle safety can be improved.

[0046] In one embodiment of the present invention, the S14 includes:

[0047] S141. Using the existing mass change measurement model, add new data to optimize the dynamic mass model, and optimize the drag coefficient by combining the latest wind speed and wind direction data through CFD simulation;

[0048] S142, based on the rolling resistance coefficient model, using new data such as tire wear, tire type, tire pressure and road surface type to make dynamic adjustments;

[0049] S143. Combine the new data from the high-precision slope sensor to optimize the slope and climbing resistance relationship model, and combine Newton's second law and vehicle dynamics parameters to construct and optimize the dynamic equation of vehicle motion;

[0050] S144, dynamically adjust the parameters in the dynamic equation according to the new data collected in real time to optimize the power demand prediction;

[0051] S145. Analyze new and old data using machine learning or pattern recognition algorithms to automatically identify driving conditions;

[0052] S146. Based on the operating condition identification results, optimize the power demand prediction model to ensure that the adaptive adjustment mechanism can reflect the latest operating conditions and external environment changes.

[0053] The working principle of the above technical solution is as follows: The model describes the mass change of the vehicle when it is loaded with different goods. By monitoring the vehicle's load in real time, the total mass of the vehicle in various states can be estimated more accurately; it is used to simulate the impact of air flow on the vehicle, especially when the vehicle is traveling at high speed. By inputting the latest wind speed and wind direction data, CFD simulation can help optimize the vehicle's air resistance coefficient (Cd value), thereby reducing unnecessary energy consumption; assuming that a logistics vehicle weighs 2 tons when empty and increases to 5 tons when fully loaded. Through the weighing sensor installed on the vehicle, the current load can be obtained in real time and the mass change model can be updated; if the vehicle is driving on the highway and encounters a crosswind with a wind speed of 20km / h and a wind direction at a 90-degree angle to the driving direction, the CFD simulation will take these conditions into account to adjust the drag coefficient to ensure a more accurate prediction; the rolling resistance coefficient model describes the friction between the tire and the ground, which affects the resistance that the vehicle needs to overcome. Rolling resistance changes with tire wear, tire pressure changes, and different road surfaces. By regularly checking tire conditions (such as wear), monitoring tire pressure, and according to different road types (such as asphalt roads, dirt roads), the rolling resistance coefficient can be dynamically adjusted to make the model closer to the actual situation. For example, a new tire has a low rolling resistance coefficient, and as the tire gradually wears, the rolling resistance increases. The system can assess the degree of wear based on the mileage of the tire or directly measure its thickness. Modern vehicles are usually equipped with a tire pressure monitoring system (TPMS), which can report the air pressure of each tire in real time. When the tire pressure is lower than the recommended value, the rolling resistance increases, the system will issue a warning and adjust the model parameters. The slope and climbing resistance relationship model reflects how the road slope affects the additional power required for the vehicle to climb. The current road slope information can be obtained in real time through the high-precision slope sensor installed on the vehicle. Newton's second law, F=ma (force equals mass multiplied by acceleration), is used to describe the relationship between the force acting on an object and its acceleration. In this case, it is used to analyze the force required to accelerate or decelerate a vehicle on different slopes; vehicle dynamics parameters include vehicle mass, air resistance coefficient, rolling resistance coefficient, etc., which together determine the vehicle's motion characteristics.By continuously updating the slope information and other relevant parameters provided by the sensors, the model can adjust the calculation of the required traction in real time to ensure that the vehicle can achieve the best performance under any road conditions; based on real-time data (such as vehicle speed, acceleration, slope, etc.), automatically adjust key parameters in the dynamic equation, such as mass, drag coefficient, etc., to reflect the latest operating conditions; by continuously updating model parameters, the accuracy of prediction of future power demand is improved to help drivers choose the most economical driving method; for example, when the vehicle enters a steep uphill slope from a flat section, the system will immediately detect the change in slope and increase the estimate of required traction accordingly; this helps to plan the engine's operating status in advance, avoid unnecessary power waste, and ensure that there is enough power to support the vehicle to climb the slope safely; machine learning / pattern recognition algorithms can learn from a large amount of historical and real-time data, identify different types of driving conditions (such as constant speed driving, acceleration, deceleration, idling, etc.), and provide the optimal power demand prediction for each condition; through continuous analysis of sensor data, the system can automatically judge The system can determine the current operating condition of the vehicle without human intervention. For example, the system can identify whether the vehicle is turning, changing lanes, or preparing to stop by analyzing data such as the vehicle's speed change rate, acceleration, and steering wheel angle. Once the driving condition is identified, the system can quickly adjust the power demand prediction model to adapt to the current driving behavior. According to the characteristics of different driving conditions, the parameters of the power demand prediction model are adjusted in a targeted manner to make it more in line with actual needs. The model has self-learning capabilities and can automatically adjust its own structure and parameters according to new operating data and external environmental changes (such as weather and traffic flow) to maintain long-term high-precision predictions. For example, on congested urban roads, where vehicles frequently start and stop, the system will adjust the model to better handle low-speed driving and frequent acceleration. On highways, more attention is paid to the power demand for high-speed and stable driving. Over time, vehicles may experience different seasons, climate conditions, or regional changes. The adaptive adjustment mechanism ensures that the model can always provide the most accurate power demand prediction, whether in hot summer or cold winter.

[0054] The effects of the above technical solutions are as follows: through the real-time updated mass change model, the estimation of vehicle load is ensured to be more accurate, thereby improving the accuracy of power demand prediction; using the latest wind speed and direction data for CFD simulation can effectively reduce the impact of air resistance and further save energy consumption; dynamically adjusting the rolling resistance coefficient according to different tire conditions and road conditions makes the model closer to the actual situation and improves the accuracy of prediction; by optimizing tire pressure and other parameters, it helps to reduce tire wear and extend its service life, indirectly reducing maintenance costs; real-time data provided by high-precision slope sensors helps optimize power demand when climbing, ensuring that the vehicle has enough power to support safe climbing while minimizing energy waste; integrating multiple dynamic parameters (such as mass, wind resistance, rolling resistance) provides a more comprehensive vehicle motion analysis tool and improves the efficiency of the overall system; the system can quickly It can quickly respond to changes in the environment and vehicle status, automatically adjust the power demand forecast, ensure that each forecast is based on the latest conditions, and improve flexibility and adaptability; through continuous optimization, it reduces unnecessary energy consumption and improves the vehicle's fuel economy and operating efficiency; it automatically identifies different driving conditions (such as constant speed driving, acceleration, deceleration, etc.) and provides the best power demand forecast for each condition without manual intervention, thus improving the level of automation; the intelligent identification and adaptive adjustment mechanism make the driving experience smoother and more efficient, and reduce the driver's operating burden; by continuously introducing new operating data and external environmental information, the model can self-learn and optimize, always keep the latest status, and adapt to the ever-changing operating environment; the model has strong adaptive capabilities and can make real-time adjustments according to actual operating conditions to ensure that each prediction reflects the latest operating conditions, thus improving the reliability and accuracy of the prediction.

[0055] In one embodiment of the present invention, the S15 includes:

[0056] S151. Optimize the response characteristic curve and efficiency curve of the power system based on the measured data through simulation testing;

[0057] S152, evaluate the adaptability of transient working conditions, combine the chemical reaction principle of hydrogen fuel cells, and optimize the efficiency model;

[0058] S153. Analyze temperature variation patterns, optimize temperature compensation mechanism and aging prediction model, and dynamically adjust hydrogen consumption calculation parameters;

[0059] S154, optimizing and matching basic power demand and power system output capacity according to real-time operation data and power system response characteristics;

[0060] S155. Establish a real-time feedback mechanism to dynamically adjust the correction strategy based on the actual operation conditions and hydrogen consumption calculation results, and continuously optimize the hydrogen consumption calculation model framework;

[0061] S156. Integrate the optimized power system response characteristic curve and efficiency curve as well as the hydrogen consumption calculation model framework into the hydrogen consumption calculation model, verify the model accuracy, and continuously optimize based on feedback.

[0062] The working principle of the above technical solution is: use computer simulation software to simulate the behavior of the power system under different working conditions (such as acceleration, deceleration, constant speed driving, etc.). These simulations can include performance evaluation of multiple components such as engines, motors, hydrogen fuel cell stacks, etc.; compare the simulation results with the data collected during actual operation, find out the differences and adjust the simulation parameters to make the simulation closer to the actual situation. The ultimate goal is to optimize the response characteristic curve (describing how the system reacts to changes in input) and efficiency curve (showing the energy conversion efficiency under different operating conditions) of the power system; for example, the change of the output power of the power system under different load conditions. If the response is too fast or too slow at low load, the control algorithm can be adjusted through simulation and measured data; such as the energy conversion efficiency of hydrogen fuel cells under different temperature and pressure conditions. By combining simulation with measured data, the trend of efficiency changes with conditions can be more accurately portrayed; study the performance of the power system under rapidly changing operating conditions (such as sudden acceleration or sudden braking). Transient conditions usually pose a challenge to the stability and efficiency of the system; consider the electrochemical reaction process inside the hydrogen fuel cell, especially when the current density changes rapidly, how it affects the battery's voltage and efficiency; based on the above analysis, improve the existing efficiency model so that it can better predict performance under transient conditions; for example, when a vehicle suddenly accelerates from idling to high-speed driving, the power system needs to quickly provide additional power; the reaction rate of hydrogen and oxygen in the fuel cell will affect the battery's output voltage and efficiency.If the reaction is too fast, it may cause local overheating or material degradation; conversely, it may reduce efficiency; study the impact of temperature on hydrogen fuel cell efficiency and identify how efficiency changes as the temperature rises or falls; automatically adjust system settings to offset the negative effects of temperature changes and maintain optimal operating efficiency; establish a model to predict the degree of aging of hydrogen fuel cells over time, taking into account factors such as material wear and chemical consumption, and adjust hydrogen consumption calculation parameters accordingly; for example, when starting in cold weather, the system automatically preheats the fuel cell to ensure that it can still work efficiently at lower temperatures; assuming that the fuel cell loses 0.5% efficiency each year, the model can Predict future efficiency declines based on known aging rates, and adjust hydrogen consumption calculation parameters in advance to maintain performance; continuously monitor vehicle speed, acceleration, slope and other information, as well as various indicators of the power system (such as voltage, current, temperature, etc.); use existing power system response characteristic curves to ensure that the system can provide the required power under various conditions; dynamically adjust the power demand prediction model through real-time data analysis to ensure that the power provided meets current needs without excessive energy consumption; for example, when the vehicle enters an uphill section, the system will immediately increase the power demand prediction based on the latest slope sensor data, while avoiding unnecessary energy waste caused by excessive acceleration; Create a closed-loop control system in which actual operating data is continuously fed back to the central processing unit to evaluate the accuracy of the model's predictions. Based on the feedback information, the parameters in the hydrogen consumption calculation model are adjusted in real time to ensure that each prediction is as close to the actual situation as possible. The model has self-learning capabilities and can gradually improve the prediction accuracy over time. For example, if the model predicts hydrogen consumption higher than the actual value, the system will analyze the reasons (such as overestimation of wind resistance) and make corresponding adjustments in subsequent predictions. If an abnormal traffic condition is encountered during a transportation process, resulting in frequent vehicle starts and stops, the system will promptly update the model parameters to adapt to this new operating condition. The optimized power The system response characteristic curve and efficiency curve as well as the hydrogen consumption calculation model framework are integrated into the hydrogen consumption calculation model to form a comprehensive prediction tool; through a series of tests (such as laboratory tests, field tests, etc.), confirm whether the model can accurately predict the hydrogen consumption; according to the test results and feedback from actual applications, further adjust and optimize the model to ensure that it always remains up to date and most accurate; for example, select several typical transportation routes for field testing, and compare the difference between the hydrogen consumption predicted by the model and the actual consumption; if it is found that the model prediction is not accurate enough in certain specific situations (such as extreme weather), the model can be improved by collecting more data under such conditions.

[0063] The effects of the above technical solutions are as follows: by testing by simulation means and optimizing based on measured data, the response characteristics and efficiency curves of the power system are ensured to be closer to the actual operation conditions, and the accuracy of hydrogen consumption prediction is improved; the power demand prediction is dynamically adjusted according to real-time operation data to ensure that the power provided not only meets the current demand but also does not consume energy excessively, reducing unnecessary energy waste; the efficiency model is optimized in combination with the chemical reaction principle of hydrogen fuel cells, so that the system can maintain high efficiency and stability under rapidly changing operating conditions (such as sudden acceleration or sudden braking), enhancing the flexibility and reliability of the system; by analyzing the temperature change law, an automatic adjustment temperature compensation mechanism is developed, and an aging prediction model is established to dynamically adjust the hydrogen consumption calculation parameters. This not only improves the working efficiency in low or high temperature environments, but also extends the service life of hydrogen fuel cells and reduces maintenance costs; a closed-loop control system is created that can dynamically adjust the correction strategy according to the actual operation conditions and hydrogen consumption calculation results to ensure that each prediction is as close to the actual situation as possible. This adaptive mechanism enables the model to continuously improve over time and always stay up to date; by continuously introducing new operating data and external environmental information, the model has the ability to self-learn, ensuring that it always reflects the latest operating conditions; the optimized power system response characteristic curve and efficiency curve and hydrogen consumption calculation model framework are integrated into the hydrogen consumption calculation model to form a comprehensive prediction tool. The reliability and practicality of the model are ensured by verifying the accuracy of the model and continuously optimizing it based on feedback.

[0064] In one embodiment of the present invention, the S2 includes:

[0065] S21. Identify environmental factors that affect hydrogen fuel cell efficiency and classify these factors based on existing models and new data;

[0066] S22. Evaluate the importance and relevance of each environmental factor, determine which factors should be considered in the model, and update the evaluation results using new data;

[0067] S23. Based on the existing test methods, test the hydrogen fuel cell under different environments and use the new data to supplement or correct the performance parameter records;

[0068] S24. Based on the new test data, the mapping relationship between environmental factors and hydrogen fuel cell efficiency is updated using statistical analysis methods;

[0069] S25. Evaluate the goodness of fit and predictive power of the preliminary model, introduce complexity assessment indicators, determine whether the model needs to be simplified or complicated, and verify these decisions with new data;

[0070] S26. Identify deficiencies in the preliminary model and design supplementary tests or collect more relevant data to enrich the training set, paying special attention to data points that can significantly improve the accuracy of the model.

[0071] S27. Apply feature engineering techniques to optimize the input feature set, and modify and retrain the model through machine learning algorithms to ensure that the model can adapt to the new training set;

[0072] S28, introduce judgment logic to guide model selection and optimization direction, set conditional branches to handle model applications under different environmental conditions, and maintain flexibility to adapt to possible changes in the future;

[0073] S29. Use adaptive mechanisms to enable the model to automatically adjust parameters or select the best version, regularly evaluate model performance, and trigger the update process when performance degradation is found, using the latest environment and performance data for retraining and verification.

[0074] The working principle of the above technical solution is as follows: First, determine which external environmental conditions will affect the efficiency of hydrogen fuel cells. These factors may include temperature, humidity, air pressure, altitude, etc.; based on existing theoretical models and newly collected data, these factors are divided into direct effects (such as temperature directly changing the battery reaction rate) and indirect effects (such as humidity indirectly affecting efficiency by affecting heat dissipation); for example, increased temperature will reduce the electrochemical reaction efficiency of hydrogen fuel cells; increased humidity may cause changes in the degree of hydration in the membrane, affecting proton conduction; indirect influencing factors include, for example, changes in air pressure may affect oxygen supply, thereby indirectly affecting battery performance; use statistical analysis or machine learning methods to evaluate the degree of influence of each environmental factor on the efficiency of hydrogen fuel cells; determine the correlation between different environmental factors, and find out those that have a significant impact on efficiency. factors; as new data is added, continuously re-evaluate which factors are most critical to ensure that the model always reflects the latest actual conditions; assume that temperature is found to be the most important factor because its changes have the most direct impact on battery efficiency; if a strong correlation is found between temperature and humidity, consider introducing interaction effects in the model; use established experimental procedures to test hydrogen fuel cells, covering various environmental conditions (such as different temperatures and humidity levels); use newly obtained test data to supplement or modify existing performance parameters so that the model more accurately reflects actual performance; for example, test the output power and efficiency of hydrogen fuel cells under low temperature (-20°C) and high temperature (40°C) conditions respectively , and use these new data to update the temperature response curve in the model; apply statistical tools such as regression analysis and principal component analysis to establish the mathematical relationship between environmental factors and battery efficiency; adjust these mapping relationships based on the latest test data to ensure that they can better predict the battery performance under different environmental conditions; find the best fit curve between temperature and battery efficiency through linear or nonlinear regression; identify which combination of environmental factors can best explain the change in battery efficiency; check the deviation between the model prediction value and the actual measurement value to evaluate the accuracy of the model; evaluate the model's predictive ability for future unknown data through methods such as cross-validation; introduce complexity indicators (such as AIC / BIC) to balance the accuracy of the model and generalization capabilities to avoid overfitting or underfitting; if a complex model performs well on the training set but poorly on the test set, consider simplifying the model structure; verify whether the simplified or complex model improves prediction accuracy by introducing new data sets; analyze areas where the model has large prediction errors and find out the causes of the errors; design additional tests for specific conditions (such as extreme temperature or humidity) or obtain more relevant data from other sources to enrich the training set; if the model does not predict accurately under high humidity conditions, more tests can be conducted under these conditions to obtain more detailed data; use historical meteorological data and field test results, especially those operating conditions that are not fully covered, such as high altitude areas;Select the features that best represent environmental factors and battery performance, remove redundant features, and create new feature combinations; use algorithms such as random forests and support vector machines to revise and retrain the model to improve its predictive ability and generalization performance; if it is found that certain features (such as temperature and humidity) contribute more to the model, retain these features and remove insignificant features; retrain the model using the enhanced feature set to ensure that it can better adapt to the new data distribution; introduce judgment logic to guide model selection and optimization direction, set conditional branches to handle model applications under different environmental conditions, while maintaining flexibility to adapt to possible future changes; provide clear rules for model selection to help decide when to use which model version or parameter configuration; automatically select the model or parameter setting that best suits the current situation based on real-time environmental conditions; ensure The system can quickly respond to new conditions or technological advances that may arise in the future; automatically switch to a low-temperature optimized model version when the temperature falls below a certain threshold; in high humidity conditions, the system will select a specially calibrated model to improve prediction accuracy; automatically adjust model parameters to ensure that they are always in the optimal state; set a periodic evaluation plan to monitor changes in model performance over time; once performance degradation is detected, immediately start the update process and retrain and verify with the latest environmental and performance data; if a sudden drop in battery efficiency is detected, the system automatically adjusts model parameters to adapt to the new operating conditions; evaluate the performance of the model once a quarter to ensure that it still meets the latest operating requirements; when the model prediction error is found to increase, immediately retrain with the latest data and deploy the updated model after verification. ;

[0075] The effects of the above technical solutions are: by carefully classifying environmental factors, the model can be ensured to fully consider all possible influencing factors. The importance assessment of environmental factors is continuously updated with new data to ensure that the model always reflects the latest actual situation; based on the new data obtained by the existing test method, the performance of hydrogen fuel cells under different environmental conditions can be more accurately described, which enhances the scientific basis of the model; the precise mapping relationship between environmental factors and battery efficiency is established through statistical analysis methods, which improves the reliability and prediction accuracy of the model; the complexity evaluation index is introduced to balance the accuracy and generalization ability of the model, avoid overfitting or underfitting, and ensure that the model is neither too simple nor too complex; by automatically adjusting parameters or selecting the best version, the model can be dynamically optimized according to the actual operating conditions to maintain long-term high-precision predictions; for the deficiencies in the preliminary model, additional tests are designed or more relevant data are collected, especially those data points that can significantly improve the accuracy of the model. For example, more tests are conducted under extreme temperature or humidity conditions to ensure that the model covers all critical operating conditions; redundant features are removed through feature engineering, new feature combinations are created, and the input feature set is optimized to improve the model's data processing capabilities and prediction accuracy; clear judgment logic and conditional branches are introduced to guide model selection and optimization direction, ensuring that the system can flexibly apply the most appropriate model version under different environmental conditions and has the ability to cope with future changes; by regularly evaluating model performance and triggering the update process, the system can respond to changes in environmental and performance data in a timely manner and always maintain the optimal state.

[0076] In one embodiment of the present invention, S3 includes:

[0077] S31. Layer the factors affecting hydrogen consumption and conduct an in-depth analysis of the factors affecting each layer based on the existing model;

[0078] S32. For each level of influencing factors, build or optimize sub-models based on existing sub-models and combined with new data, and integrate these sub-models through Bayesian networks;

[0079] S33. Conduct preliminary analysis and evaluation of the comprehensive hydrogen consumption model, use historical data to backtest the effectiveness of the model, and dynamically adjust the model based on new data;

[0080] S34, during the model prediction process, automatically switching to the sub-model that best suits the current situation through conditional judgment logic;

[0081] S35. Dynamically adjust the model parameters or structure according to the real-time data and prediction results, establish a feedback mechanism to compare the actual hydrogen consumption data with the predicted value, identify the source of error, and optimize the model accordingly;

[0082] S36. Use particle swarm optimization and other algorithms to further optimize the model, introduce new influencing factors and characteristic variables to improve the model structure;

[0083] S37. Use independent test sets to validate the optimized model, and regularly update and improve the model to ensure that it continues to reflect actual conditions.

[0084] The working principle of the above technical solution is: divide all factors that may affect the hydrogen consumption of hydrogen fuel cell vehicles into multiple levels, such as basic power requirements, environmental conditions, driving behavior, etc.; use existing theoretical models and simulation tools to conduct detailed research on the influencing factors at each level to find out how they affect hydrogen consumption individually and together.

[0085] Layered example:

[0086] The first layer (basic power requirements): includes vehicle mass, speed, acceleration, etc.

[0087] The second layer (environmental conditions): temperature, humidity, altitude, etc.

[0088] The third level (driving behavior): sudden acceleration, frequent braking, etc.

[0089] For example, in the first layer, the energy required by the vehicle under different load conditions is analyzed; in the second layer, the impact of temperature changes on battery efficiency is studied; and in the third layer, the additional energy consumption under different driving habits is evaluated.

[0090] According to the specific influencing factors at each level, the sub-model is constructed or optimized using the existing model framework and combined with the latest measured data; the Bayesian network is used as an integration tool to connect the various sub-models to form an overall multi-level model. The Bayesian network can handle uncertainty and provide probabilistic predictions; for sub-model construction, for example, for the basic power demand layer, a kinetic model based on Newton's second law can be constructed; for the environmental conditions layer, a thermodynamic model that considers factors such as temperature and humidity is constructed; assuming that both temperature and humidity affect battery efficiency, the Bayesian network can represent the relationship between the two and how they jointly affect the final hydrogen consumption.This network structure allows the model to make more reasonable predictions under uncertain conditions; conduct preliminary tests on the integrated multi-level model to check its logical correctness and the rationality of the calculation results; verify the performance of the model with past data sets to ensure that it can give accurate predictions under known conditions; automatically update model parameters or structures as new operating data continues to flow in to adapt to changing actual conditions; for example, select actual operating data over a certain period of time, compare the difference between the model-predicted hydrogen consumption and the actual consumption, and evaluate the accuracy of the model; if it is found that the model prediction is not accurate enough under certain specific conditions (such as extreme weather), it can be improved by collecting more data under such conditions. data to improve the model; set a series of rules or conditions, and when specific conditions are met, the system will automatically select the most appropriate sub-model to make predictions; ensure that the model can flexibly select the best prediction solution based on the real-time environment and operating conditions to improve prediction accuracy; for example, when the temperature is below a certain threshold, the system will automatically switch to a low-temperature optimized sub-model; or when frequent starts and stops are detected, switch to a sub-model suitable for urban congested road conditions; based on real-time collected data and prediction results, continuously adjust model parameters or structures to keep the model up to date; create a closed-loop control system to compare actual operating data with predicted values, identify sources of error, and optimize accordingly model; for example, if the predicted value is found to be higher than the actual consumption, the system will analyze the reasons (such as the wind resistance is estimated too high) and make corresponding adjustments in subsequent predictions; assuming that an abnormal traffic condition is encountered during a transportation process, resulting in frequent vehicle starts and stops, the system will update the model parameters in a timely manner to adapt to this new working condition; Particle Swarm Optimization (PSO) is an optimization algorithm based on swarm intelligence, which is used to find the best combination of model parameters to minimize errors or maximize performance indicators; According to new discoveries or needs in actual applications, more influencing factors and characteristic variables are introduced to make the model more comprehensive and accurate; For example, the PSO algorithm is used to optimize the weights of various parameters in the model to make The prediction results should be as close to the actual values ​​as possible; assuming that it is found that the road slope has a significant impact on hydrogen consumption, the slope sensor data can be added to the model to more accurately reflect the actual operating conditions; the optimized model is verified using an independent data set that is not involved in the training to ensure its generalization ability; a periodic evaluation plan is set to monitor the changes in model performance over time, and update and improve it when necessary to maintain the long-term effectiveness of the model; for example, several typical transportation routes are selected for field testing, and the difference between the hydrogen consumption predicted by the model and the actual consumption is compared to confirm the accuracy of the model; the performance of the model is evaluated once a quarter to ensure that it still meets the latest operational requirements. If the model prediction error is found to increase, retrain it immediately using the latest data and deploy the updated model after verification.

[0091] The effects of the above technical solution are as follows: by carefully classifying the factors affecting hydrogen consumption and deeply analyzing the factors affecting each layer, it is ensured that the model can comprehensively consider all possible factors and improve the precision of the prediction; the sub-models at all levels are integrated using the Bayesian network, which not only enhances the scientificity and logic of the model, but also can handle uncertainty and complex relationships, and improve the accuracy of the prediction; by using historical data to backtest the effectiveness of the model, the performance of the model under known conditions is verified, and its reliability and credibility are enhanced; the optimized model is regularly verified using independent test sets to ensure that it has good generalization ability and can make accurate predictions under unknown conditions; the conditional judgment logic is automatically used to Select the sub-model that best suits the current situation, so that the system can flexibly apply the most appropriate prediction scheme under different working conditions, thereby improving the adaptability and response speed of the system; dynamically adjust the model parameters or structure according to real-time data and prediction results, establish a feedback mechanism to identify the source of errors, and optimize the model accordingly to ensure that the system is always in the optimal state; introduce advanced algorithms such as particle swarm optimization to further optimize the model, and continuously introduce new influencing factors and characteristic variables to improve the model structure to ensure that the model can continue to improve with the advancement of new data and technology; set up a periodic evaluation plan to monitor changes in model performance over time, and update and improve it when necessary to ensure that the model always remains up to date and accurate.

[0092] In one embodiment of the present invention, the S32 includes:

[0093] S321. Based on the vehicle status, use high-precision sensor data to build a vehicle physical characteristics model, use machine learning algorithms to identify different driving styles based on driving behavior data, and establish a quantitative model of the impact of driving behavior on hydrogen consumption;

[0094] S322. Combine the real-time operation data of the vehicle and the working characteristics of the hydrogen fuel cell to build a dynamic efficiency model, integrate the vehicle state model, driving behavior model and hydrogen fuel cell efficiency model to form a hydrogen consumption prediction model at the micro level;

[0095] S323. Analyze the road conditions, traffic flow, congestion, etc. of different routes using GIS data and real-time traffic monitoring data, build a route selection model, and evaluate the impact of each route on hydrogen consumption;

[0096] S324. Establish a correlation model between road type and vehicle energy consumption based on the geometric characteristics and traffic features of the road type; construct a hydrogen consumption optimization model at the meso-level based on route selection and road type analysis combined with real-time vehicle operation data;

[0097] S325. Collect and analyze regional climate data and policy environment, and evaluate the indirect impact of these factors on hydrogen consumption; based on historical data and macroeconomic indicators, use time series analysis to predict the possible impact of macroeconomic factors on hydrogen consumption in the future;

[0098] S326. Convert the results of regional climate and policy analysis into quantifiable parameters, construct a quantitative model of hydrogen consumption impact at the macro level, and provide macro-level input for the comprehensive hydrogen consumption model;

[0099] S327. Design the Bayesian network structure based on the logical relationship between the micro, meso and macro sub-models, and clarify the dependency and causal relationship between each node;

[0100] S328. Based on historical data and expert knowledge base, the prior probability and conditional probability are set for each node in the Bayesian network. The prediction results of sub-models at all levels are integrated into the comprehensive hydrogen consumption model using the Bayesian network inference algorithm to achieve the fusion and comprehensive analysis of multi-source data.

[0101] The working principle of the above technical solution is as follows: collect physical parameters (such as speed, acceleration, slope, etc.) during vehicle operation through high-precision sensors (such as accelerometers, gyroscopes, etc.) to build a mathematical model that describes the dynamic characteristics of the vehicle; use machine learning algorithms (such as decision trees, random forests, deep learning, etc.) to analyze driving behavior data (such as accelerator pedal position, braking frequency, etc.) and identify different driving styles (such as aggressive, economical, etc.); establish a quantitative relationship between driving behavior and hydrogen consumption, and evaluate the specific impact of different driving styles on hydrogen consumption; for example, based on the changes in the speed and acceleration of the vehicle, predict the changes in its kinetic energy and potential energy, and then calculate the energy consumption; assuming that by analyzing the driving data, it is found that a driver frequently accelerates and brakes suddenly, the system can identify that this is an aggressive driving style; studies have shown that aggressive driving styles consume about 20% more hydrogen fuel than mild driving styles on average; considering the real-time operation data of the vehicle (such as speed, load, etc.) and hydrogen consumption The working characteristics of the fuel cell (such as power output, efficiency curve, etc.) are used to build a dynamic model that can reflect the changes in vehicle efficiency under different working conditions; the vehicle state model, driving behavior model and hydrogen fuel cell efficiency model are integrated to form a complete hydrogen consumption prediction model at the micro level; for example, hydrogen fuel cells may be more efficient when driving at high speeds; while at low speeds or when starting and stopping frequently, the efficiency will decrease; assuming that the vehicle is driving at a constant speed on a flat road and the driver adopts an economical driving style, the integrated model can be used to accurately predict the hydrogen consumption of the section; using geographic information system (GIS) data and real-time traffic monitoring data, analyze factors such as road conditions (such as slope, curvature), traffic flow, congestion, etc. of different routes; evaluate the specific impact of these factors on hydrogen consumption, and provide a basis for selecting the optimal transportation route; for example, some routes are shorter but often congested, while other longer routes are relatively unobstructed.Based on this information, the system can recommend the most energy-efficient route; suppose a route passes through multiple traffic lights, resulting in frequent starts and stops, which will significantly increase hydrogen consumption; while another route is longer, but has stable traffic and lower overall hydrogen consumption; based on the geometric characteristics of the road (such as width, curve radius, etc.) and traffic characteristics (such as speed limit, number of lanes, etc.), a correlation model between road type and vehicle energy consumption is established; combined with the route selection and road type analysis results, as well as real-time vehicle operation data, a hydrogen consumption optimization model at the meso level is constructed to further optimize the choice of transportation routes; for example, highways usually have higher speed limits and fewer traffic lights, so compared with urban roads, vehicles consume less hydrogen when traveling on highways; assuming that a transportation task needs to cross mountainous areas, the system can adjust the vehicle speed according to the characteristics of the mountain roads, optimize driving behavior, and thus reduce hydrogen consumption; collect climate data (such as temperature, humidity, etc.) and policy environment (such as emission standards, subsidy policies, etc.) between regions to evaluate their indirect impact on hydrogen consumption; based on historical data Based on the data of the global climate and macroeconomic indicators, time series analysis methods (such as ARIMA, LSTM, etc.) are used to predict the possible impact of macro factors on hydrogen consumption in the future. For example, the efficiency of hydrogen fuel cells will decrease under low temperature conditions, and environmental protection policies in some regions may prompt companies to choose more efficient vehicles or routes. Assuming that the temperature in a certain quarter in the future is predicted to be abnormally low based on the data of the past few years, the system can adjust the vehicle configuration or route planning in advance to cope with the potential increase in hydrogen consumption. The impact of climate conditions and policy environment is converted into specific numerical parameters, such as temperature correction coefficients, policy incentive factors, etc. A quantitative model of hydrogen consumption impact at the macro level is constructed as one of the inputs of the comprehensive hydrogen consumption model to help comprehensively evaluate the impact of macro factors on hydrogen consumption. For example, for every 1°C decrease, the efficiency of hydrogen fuel cells decreases by 0.5%, which can be converted into a temperature correction coefficient. Assuming that a new subsidy policy is introduced to encourage the use of clean energy vehicles, the system can quantify the impact of this policy on hydrogen consumption and incorporate it into the macro-level model.Design a Bayesian network structure to integrate sub-models at the micro, meso and macro levels; clarify the dependencies and causal relationships between nodes to ensure that the model can correctly reflect the interactions between different factors; for example, construct a Bayesian network containing multiple nodes such as vehicle status, driving behavior, road type, and climate conditions; assume that driving behavior affects vehicle status, which in turn affects hydrogen consumption, and these causal relationships can be clearly expressed in the Bayesian network; set reasonable prior probabilities and conditional probabilities for each node in the Bayesian network based on historical data and expert knowledge base; use the Bayesian network inference algorithm to integrate the prediction results of sub-models at each level into the comprehensive hydrogen consumption model to achieve the fusion and comprehensive analysis of multi-source data; for example, based on past experience, set the probability of a specific driving behavior occurring on busy urban roads to 0.7; assume that the current weather is cold , the system can infer the possibility of increased hydrogen consumption in this case through the Bayesian network and adjust the transportation plan accordingly; build a vehicle physical characteristics model through high-precision sensor data, identify the impact of driving behavior on hydrogen consumption, and combine the working characteristics of hydrogen fuel cells to form a hydrogen consumption prediction model at the micro level; use GIS data and real-time traffic monitoring data to analyze the road conditions and traffic flow of different routes, build a route selection model, and combine the relationship between road type and vehicle energy consumption to optimize transportation route selection; analyze the indirect impact of climate data and policy environment on hydrogen consumption, predict the changing trend of future macro factors, and quantify these factors into parameters as input to the comprehensive hydrogen consumption model; design the Bayesian network structure, clarify the dependency and causal relationship between sub-models at all levels, use the Bayesian network reasoning algorithm to integrate the prediction results at all levels, and realize the fusion and comprehensive analysis of multi-source data.

[0102] The effects of the above technical solutions are as follows: constructing a vehicle physical characteristics model through high-precision sensor data, identifying the impact of driving behavior on hydrogen consumption, and combining the working characteristics of hydrogen fuel cells to form a hydrogen consumption prediction model at the micro level; improving the accuracy of hydrogen consumption prediction for a single vehicle under specific driving conditions, providing a scientific basis for personalized management and optimization; using GIS data and real-time traffic monitoring data to analyze the road conditions and traffic flow of different routes, constructing a route selection model, and optimizing the transportation route selection by combining the relationship between road type and vehicle energy consumption; optimizing the transportation route selection, reducing hydrogen consumption, improving operational efficiency, reducing unnecessary fuel waste, and directly reducing operating costs; analyzing the indirect impact of climate data and policy environment on hydrogen consumption, predicting the changing trend of future macro factors, and quantifying these factors into parameters as inputs to the comprehensive hydrogen consumption model; enhancing the system's adaptability and long-term reliability, enabling it to respond to changing actual conditions in a timely manner, always maintaining the optimal state, and avoiding prediction deviations caused by external factors; more accurate hydrogen consumption estimation and optimization management help reduce energy consumption, indirectly reduce operating costs, and improve the company's profitability; by optimizing transportation routes and hydrogen consumption estimation, fuel consumption is reduced, and further The system has further reduced operating costs and improved economic benefits; more accurate hydrogen consumption estimation and optimization management help reduce energy consumption and greenhouse gas emissions, meet the goal of sustainable development, and promote the development of green logistics; reduced carbon footprint, met environmental protection regulations, and enhanced the company's social responsibility and social image; provided scientific decision support tools, simplified operating procedures, improved work efficiency and service quality, and enhanced user satisfaction; the intelligent decision support system improved driving safety and comfort, while quickly responding to changes in customer needs, improving service quality and customer satisfaction; by optimizing transportation routes and hydrogen consumption estimation, it improved resource utilization efficiency and enhanced the company's competitiveness in the market; efficient operation and low-cost advantages enable the company to stand out in the fiercely competitive market, win more business opportunities, and enhance market competitiveness; the Bayesian network structure is designed according to the logical relationship between sub-models at all levels, the dependency and causal relationship are clarified, and the prior probability and conditional probability are set based on historical data and expert knowledge base, and the Bayesian reasoning algorithm is used to integrate the prediction results at all levels; the fusion and comprehensive analysis of multi-source data are realized, ensuring that the model can fully consider the influence of various factors and improve the accuracy and reliability of the prediction.

[0103] In one embodiment of the present invention, the S4 includes:

[0104] S41. Plan multiple transport routes, consider the impact of traffic conditions at different time periods, and optimize them by using existing traffic prediction models and adding new data;

[0105] S42. Based on historical meteorological data and real-time weather forecasts, the transportation environment is simulated by simulation software to ensure that the simulation conditions are as close to the actual conditions as possible;

[0106] S43, setting vehicle parameters for a specific vehicle model, inputting corresponding parameters of the hydrogen fuel cell, and optimizing using the existing vehicle dynamics model combined with the new data;

[0107] S44, simulating different time and route conditions based on real-time traffic data and historical statistical data, and using cloud computing resources to perform simulation calculations of multiple routes in parallel;

[0108] S45. Record key data in the simulation process in real time, output simulation results, and use these results for subsequent model optimization and verification.

[0109] The working principle of the above technical solution is as follows: based on the starting point and the end point, multiple possible transportation routes are generated. These routes can include different types of sections such as highways and urban roads; considering the changes in traffic flow at different times of the day (such as peak hours and non-peak hours), the existing traffic prediction model is used to evaluate the traffic efficiency of each time period; combined with the latest real-time traffic data (such as from GPS, traffic cameras, intelligent transportation systems, etc.), the traffic prediction model is updated and optimized to ensure that it can reflect the latest traffic conditions; for example, suppose there are multiple route options from City A to City B, one mainly through the highway, and the other more through urban roads; between 8 am and 10 am, some sections may experience severe traffic congestion, so when planning, those routes that are relatively unobstructed during this period will be given priority; if a section is under construction or a traffic accident occurs, real-time traffic data can help adjust route planning to avoid unnecessary delays; analyze meteorological data in the past few years to understand the weather patterns of different seasons, months and even specific dates; incorporate the latest weather forecast information (such as temperature, humidity, precipitation probability, etc.) into the simulation environment to provide High simulation authenticity; Use professional simulation software to create a virtual transportation environment, simulate various meteorological conditions during vehicle driving, and ensure that the simulation results are as close to reality as possible; for example, there may be risks of snowfall or ice in winter, and high temperatures or heavy rain in summer, all of which need to be taken into account in the simulation; If the forecast shows that there will be heavy rain on a certain day, the corresponding road slippery coefficient can be set in the simulation to affect the rolling resistance and braking performance of the vehicle; Use simulation software to simulate the scene of vehicle driving on rainy days, including factors such as reduced visibility and reduced road friction, making the simulation more realistic; Set the basic parameters of the vehicle (such as mass, size, air resistance coefficient, etc.) according to the actual vehicle model (such as heavy trucks, light trucks, etc.); Input parameters related to hydrogen fuel cells (such as power output, efficiency curve, operating temperature range, etc.) to ensure that the simulation can accurately reflect the behavior of this type of battery; Use existing vehicle dynamics models and combine them with the latest measured data (such as slope sensors, speed sensors, etc.) to optimize model parameters and improve simulation accuracy.For example, for a specific model of hydrogen fuel cell logistics vehicle, its empty mass is set to 5 tons, its full mass is set to 10 tons, and its frontal area is set to 7 square meters; it is assumed that the maximum output power of the hydrogen fuel cell of this model at room temperature is 150kW, the efficiency is about 60%, and the efficiency decreases as the temperature rises; by combining the latest collected speed, acceleration, slope and other data, the parameters in the model are dynamically adjusted to ensure that it always reflects the latest operating status; comprehensive real-time traffic data (such as current road conditions, accident information, etc.) and historical statistical data (such as average vehicle speed, congestion frequency, etc.) ), create detailed simulation scenarios for each route; simulate traffic conditions in different time periods (such as morning and evening peak hours, off-peak hours) for multiple alternative routes, and evaluate the traffic efficiency and hydrogen consumption of each route; with the powerful computing power of the cloud computing platform, execute multiple simulation tasks in parallel, greatly shortening the simulation calculation time and improving work efficiency; for example, during the morning peak hour, the historical average speed of a route is 30km / h, but currently due to a traffic accident, the speed drops to 15km / h, and these factors will be taken into account during the simulation; assuming that there are three routes to choose from, Simulations were performed at 7 a.m., 12 p.m., and 7 p.m. to compare their travel efficiency and hydrogen consumption in different time periods. The three simulation tasks were run simultaneously on the cloud platform to obtain simulation results for all routes in a short period of time and to quickly make the best choice. During the simulation, important operating parameters (such as speed, acceleration, hydrogen consumption, etc.) and various operating conditions encountered (such as climbing, sudden braking, etc.) were recorded in real time. The data obtained from the simulation was organized into a report format and provided to decision makers for reference to help them choose the most appropriate transportation route. Feedback information from the simulation results was used to identify deficiencies in the model and improve it accordingly. At the same time, new simulation results were used to verify the accuracy of the model to ensure its continuous optimization. For example, the instantaneous power output of the hydrogen fuel cell, the average speed of the vehicle, and the total hydrogen consumption were recorded during each simulation. A detailed simulation report was generated, including information such as the estimated driving time, total hydrogen consumption, and potential risk points for each route. If it was found that the hydrogen consumption predicted by the model was too high under a certain condition (such as extreme low temperature), the model could be improved by collecting more data under such conditions and its accuracy could be verified with new simulation results.

[0110] The effects of the above technical solutions are as follows: by planning multiple transportation routes and considering traffic conditions in different time periods, logistics companies can choose the most efficient route to reduce transportation time and fuel consumption; by using the powerful computing power of cloud computing to perform simulation calculations of multiple routes in parallel, the simulation time is greatly shortened and the decision-making speed is improved; the traffic prediction model is updated in combination with new data to ensure that it can reflect the latest traffic conditions and improve the accuracy and reliability of route selection; a simulation environment close to the actual situation is constructed through historical meteorological data and real-time weather forecasts, making the simulation results more credible and reducing the probability of accidents; using simulation software to simulate the transportation environment under various weather conditions, identifying potential risk points in advance, such as slippery roads, bad weather, etc., to enhance driving safety; setting detailed vehicle parameters for specific models and inputting hydrogen fuel cells relevant parameters to ensure that the simulation can accurately reflect the actual vehicle performance and improve the reliability of operation; the existing vehicle dynamics model is used in combination with new data to optimize the parameters of the hydrogen fuel cell, which improves the accuracy of hydrogen consumption estimation and reduces unnecessary fuel waste; the optimal route is selected according to the simulation results to avoid congested sections or high-energy consumption conditions, which directly reduces operating costs; more accurate hydrogen consumption estimation and optimization management help reduce energy consumption and greenhouse gas emissions, meet the goals of sustainable development, and promote the development of green logistics; scientific decision-making support is provided through simulation calculations, which simplifies the operating process, improves driving safety and comfort, and enhances user satisfaction; key data in the simulation process is recorded in real time, simulation results are output, and these results are used for subsequent model optimization and verification to ensure the long-term reliability and continuous improvement capabilities of the system.

[0111] In one embodiment of the present invention, the S44 includes:

[0112] S441, integrating real-time traffic data from multiple channels; and pre-processing the acquired real-time traffic data;

[0113] S442, analyzing historical traffic data, using data mining algorithms to extract typical patterns and rules of traffic flow and congestion from historical data; combining historical statistical data with current real-time data, using time series analysis algorithms to predict traffic conditions in the future;

[0114] S443. According to the reliability of real-time data and the stability of historical data, different weights are assigned to the two to build a dynamic fusion model; based on real-time traffic data and historical prediction results, multiple traffic scenarios are constructed and the probability of each scenario is evaluated;

[0115] S444. According to the constructed scenario, different simulation scenarios are set for each transportation route, and computing resources are dynamically allocated according to the complexity and computing requirements of the simulation task using the cloud computing platform; and simulation calculations of multiple transportation routes are performed simultaneously through parallel algorithms;

[0116] S445. During the simulation process, the calculation progress and result quality are monitored in real time, the simulation parameters are fine-tuned using heuristic search, and key data of the simulation process are recorded in real time;

[0117] S446. Summarize the simulation results of each transport route to form a comprehensive evaluation report; conduct in-depth analysis of the simulation results, identify potential problems and improvement points, and put forward targeted suggestions and optimization plans.

[0118] The working principle of the above technical solution is as follows: collect real-time traffic data from different sources (such as traffic cameras, GPS devices, mobile applications, intelligent transportation systems, etc.); integrate these scattered data sources to ensure that all data can be processed and analyzed on a unified platform; for example, obtain real-time traffic information of urban roads through the API provided by the traffic management department, and use the vehicle's own GPS device to collect the speed and location information of each vehicle; centralize these data into a data center for subsequent processing and analysis; remove noise in the data (such as duplicate records, erroneous values ​​or outliers) to improve data quality; standardize data from different sources and formats into a unified format and unit for subsequent processing; for example, remove Abnormal records of speeds many times higher than the speed limit, or merging repeated location records at the same time point; reviewing traffic data over a period of time in the past to identify common traffic patterns and rules; using data mining techniques such as cluster analysis and association rule learning to identify features such as traffic flow peaks and areas with frequent congestion; for example, analyzing the changes in traffic flow between 5pm and 7pm every Friday in the past year, it was found that this period is a typical rush hour after work, which is prone to traffic congestion; using the K-means clustering algorithm to find the most congested sections of road, or using association rule learning to find out which factors (such as weather conditions, holidays) will lead to increased traffic flow; combining the trends and periodicity of historical data with current real-time data to predict future short-term traffic jams. medium-term traffic conditions; use time series prediction models such as ARIMA and LSTM to generate estimates of future traffic conditions; for example, based on the average vehicle speed from 8 to 9 a.m. every day in the past week and the current real-time vehicle speed, predict the changes in traffic flow in the next hour; assuming the use of an LSTM neural network model, the model can accurately predict future traffic conditions based on learning from historical and real-time data; evaluate the reliability and stability of real-time data and historical data, and assign appropriate weights to both; build a fusion model that comprehensively considers the impact of real-time and historical data to provide more accurate traffic condition predictions; for example, during peak traffic hours, real-time data may be more important, so it is given a higher weight; while during non-peak hours, historical data may be more important. Historical data may be more valuable for reference, and its weight can be appropriately increased; assuming that in a certain time period, real-time traffic data shows that an accident has occurred on a certain road, at this time, more reliance should be placed on real-time data for prediction; in other time periods, more reliance should be placed on historical data; based on real-time traffic data and historical prediction results, different traffic scenarios (such as normal traffic, slight congestion, severe congestion, etc.) can be simulated; Bayesian inference or other statistical methods can be used to evaluate the probability of each scenario occurring; for example, assuming that current real-time data shows that the traffic volume on a certain road is gradually increasing, two scenarios of "slight congestion" and "severe congestion" can be constructed; Bayesian inference can be used to calculate the probability of these two scenarios occurring in the next half hour, helping decision makers make more informed choices;According to different traffic scenarios, specific simulation parameters (such as estimated driving time, hydrogen consumption, etc.) are set for each transport route; according to the complexity of the simulation task and the computing requirements, the computing resources of the cloud computing platform are dynamically allocated to ensure efficient execution; through parallel computing technology, the simulation calculations of multiple transport routes are carried out simultaneously, which greatly shortens the simulation time; for example, for the "slight congestion" scenario, the estimated driving time of a transport route is set to 1.5 hours and the hydrogen consumption is set to 10kgH; 2 . Assuming that a simulation task involves a large amount of data processing, the system will automatically allocate more computing nodes to speed up the calculation; through a distributed computing framework (such as Apache Spark), simulation tasks for multiple routes can be run simultaneously, improving work efficiency; during the simulation process, the calculation progress and result quality are continuously monitored to ensure the smooth progress of the simulation process; heuristic search algorithms are used to optimize simulation parameters to improve the accuracy and efficiency of the simulation; key data in the simulation process, such as simulation parameter adjustments, calculation progress, etc., are recorded to provide support for subsequent analysis; for example, the simulation progress bar of each route can be viewed in real time to ensure that no simulation tasks are stuck or failed; assuming that the simulation results show that the hydrogen consumption of a certain route is too high, the speed or acceleration parameters of the vehicle can be adjusted through a heuristic search algorithm to optimize hydrogen consumption; record The time point and specific value of each parameter adjustment, as well as other key data in the simulation process, are convenient for subsequent analysis and report generation; the simulation results of all transportation routes are summarized to generate a detailed comprehensive evaluation report; the simulation results are analyzed in detail to identify potential problems and improvement points; based on the analysis results, targeted improvement suggestions and optimization plans are proposed to help logistics companies optimize their operating strategies; for example, a report containing the estimated driving time, hydrogen consumption, potential risk points, etc. for each route is generated; assuming that a certain route often encounters traffic jams during a specific time period, the analysis may be due to school dismissal or shopping mall promotions nearby; it is recommended to avoid these time periods or choose alternative routes, and at the same time, plans to optimize fleet scheduling and driving behavior are proposed to reduce hydrogen consumption and improve efficiency.

[0119] The effects of the above technical solutions are as follows: the above technical solutions improve the quality and consistency of data, ensure the accuracy of subsequent analysis and prediction, and reduce misjudgments caused by data problems; through comprehensive analysis of historical data and real-time data, the accuracy and reliability of traffic condition prediction are improved, helping logistics companies to prepare in advance and avoid potential risks; the adaptability and long-term reliability of the model are maintained, so that it can respond to the actual situation of changes in a timely manner, always maintain the optimal state, and avoid prediction deviations caused by model aging; by optimizing transportation routes and hydrogen consumption estimates, fuel consumption is reduced, operating costs are further reduced, and economic benefits are improved; energy consumption and greenhouse gas emissions are reduced, which is in line with the goal of sustainable development, promotes the development of green logistics, and enhances the social responsibility and social image of enterprises; the intelligent decision support system improves driving safety and comfort, while quickly responding to changes in customer needs, improving service quality and customer satisfaction; efficient operation and low-cost advantages enable companies to stand out in the fiercely competitive market, win more business opportunities, and enhance market competitiveness; ensure the efficiency of the simulation process and the accuracy of the results, provide more refined adjustment methods, and improve the scientificity and rationality of the final decision; provide comprehensive evaluation and improvement suggestions to help logistics companies continuously optimize their operating strategies and ensure the long-term effectiveness and continuous improvement capabilities of the system.

[0120] In one embodiment of the present invention, S5 includes:

[0121] S51. Clarify the optimization goals, set weights based on the enterprise operation strategy and market demand, and build a multi-objective optimization function;

[0122] S52. Evaluate the applicability of different multi-objective optimization algorithms, select the most appropriate algorithm, and adjust algorithm parameters based on new data;

[0123] S53. Use the simulation results as input to run the multi-objective optimization algorithm, generate the Pareto frontier solution set, analyze the performance of different solutions, and select the best transportation route from the solution set, while using new data to continuously optimize the selection process.

[0124] The working principle of the above technical solution is as follows: determine multiple goals that need to be optimized simultaneously, such as minimizing hydrogen consumption, reducing transportation time, reducing operating costs, etc.; assign appropriate weights to each goal according to the company's operating strategy (such as giving priority to cost control or delivery speed) and market demand (such as the time window of customer demand); integrate multiple goals into a mathematical function and construct a multi-objective optimization model through weighted summation or other methods; compare multiple multi-objective optimization algorithms (such as genetic algorithms, particle swarm optimization, NSGA-II, etc.) and evaluate their performance in dealing with specific problems; select the algorithm that best suits the current problem based on the evaluation results; dynamically adjust algorithm parameters as new data is introduced to maintain optimal performance; for example, for problems with complex constraints, genetic algorithms may perform better; while for continuous optimization problems, particle swarm optimization may be a better choice; assuming that after evaluation, it is found that the NSGA-II algorithm can better handle multi-objective optimization problems and the generated Pareto frontier solution set is of high quality, so this algorithm is selected; as more measured data is collected, it may be found that certain parameters (such as population size, crossover probability, etc.) need to be adjusted to obtain better results; the simulation results obtained in S4 (such as hydrogen consumption of different routes, estimated driving time, etc.) are used as Provide the input data to the multi-objective optimization algorithm; calculate a series of non-inferior solutions (i.e., Pareto optimal solutions) through the multi-objective optimization algorithm, which represent the trade-off relationship between different objectives; evaluate the performance of each Pareto solution, understand their advantages and disadvantages in different objectives, and help decision makers make choices; select one or more optimal transportation routes from the Pareto frontier solution set based on the specific needs and preferences of the enterprise; continuously update and optimize the selection process as new simulation results and actual operation data accumulate to ensure that the most appropriate route is always selected; for example, input the hydrogen consumption, expected driving time and potential risks of each route points and other information; suppose a set of different transportation routes are obtained, some of which have low hydrogen consumption but long transportation time, while others have short transportation time but high hydrogen consumption; analyze the performance of these solutions on different objectives, for example, some routes have slightly higher hydrogen consumption but can significantly shorten transportation time, which is suitable for urgent orders; while other routes are more suitable for routine transportation tasks; select the most suitable route from the Pareto frontier solution set based on the company's current needs (such as whether cost or time is prioritized); for example, if a route is found to often encounter traffic congestion during a specific time period, the selection process can be improved by collecting more real-time traffic data to avoid similar situations from happening again.

[0125] The effects of the above technical solutions are as follows: by determining multiple goals that need to be optimized simultaneously and setting weights according to the enterprise's operating strategy and market demand, a more scientific and reasonable multi-objective optimization function is constructed; for example, if a logistics company pays more attention to cost control, it can give higher weights to "minimizing hydrogen consumption" and "reducing operating costs"; if the customer has strict requirements on delivery time, the weight of "reducing transportation time" will be higher; by evaluating the applicability of different multi-objective optimization algorithms, the algorithm that best suits the current problem is selected, and the algorithm parameters are dynamically adjusted according to new data to ensure that the algorithm is always in the optimal state, thereby improving the optimization accuracy and adaptability; running the multi-objective optimization algorithm with simulation results as input to generate a series of non-inferior solutions (i.e., Pareto optimal solutions), which represent the trade-off relationship between different objectives and provide more flexibility in selection; evaluating the performance of each Pareto solution, understanding their advantages and disadvantages in different objectives, helping decision makers make the most appropriate route selection, and enhancing the choice Reliability and rationality; through the hydrogen consumption minimization goal in the multi-objective optimization function, accurate prediction and optimization management of hydrogen consumption are achieved, which reduces unnecessary fuel waste and directly reduces operating costs; the best transportation route is selected from the Pareto frontier solution set to avoid congested sections or high-energy consumption conditions, further reducing operating costs; more accurate hydrogen consumption prediction and optimization management help reduce energy consumption and greenhouse gas emissions, meet the goals of sustainable development, and promote the development of green logistics; provide scientific decision support tools, simplify operating procedures, improve driving safety and comfort, and enhance user satisfaction; through continuous optimization of the selection process, it can quickly respond to changes in customer needs, improve service quality and customer satisfaction; use simulation results and actual operation data to continuously optimize the selection process to ensure the long-term reliability and continuous improvement capabilities of the system, so that the model always reflects the latest actual situation; with the introduction of new data, dynamically adjust the algorithm parameters to maintain the efficiency and accuracy of the optimization algorithm.

[0126] In one embodiment of the present invention, S6 includes:

[0127] S61. Establish a real-time data collection system to continuously monitor key data during vehicle operation, ensure real-time transmission and monitoring between the vehicle and the data center, and use encryption technology and security protocols to ensure data security;

[0128] S62. Compare the actual operation data with the simulation results, evaluate the model accuracy, identify the source of error and analyze the cause, and dynamically correct the model using the online learning algorithm;

[0129] S63. Dynamically revise the comprehensive hydrogen consumption model based on real-time data, and regularly evaluate the performance of the revised model to ensure that the model is always in the optimal state;

[0130] S64. Re-perform simulation calculations and optimization analysis based on the revised model to ensure that the selection of transportation routes and hydrogen consumption estimates are always maintained at the optimal level.

[0131] The working principle of the above technical solution is as follows: install sensors and other monitoring devices (such as GPS, speed sensors, acceleration sensors, temperature sensors, etc.) to continuously collect key data during vehicle operation; use the Internet of Things (IoT) technology to achieve real-time data transmission between the vehicle and the data center to ensure that the data can be reached and processed immediately; use encryption technology and security protocols (such as SSL / TLS, AES encryption, etc.) to protect the security of data during transmission and storage to prevent data leakage or tampering; for example, sensors installed on the vehicle can record information such as speed, acceleration, position, battery temperature, etc. once a second; use 4G / 5G networks or dedicated communication links to send this data to the data center in real time, and the data center can immediately analyze and respond to any abnormal situations; all transmitted data is encrypted with AES-256 to ensure that sensitive information is protected even in cyber attacks; compare the actual operation data collected in real time with the results of previous simulation calculations to find out the differences between the two; evaluate the accuracy of the existing model through comparative analysis, identify the source of error, and analyze the cause of the error; apply online learning algorithms ( such as incremental learning, adaptive filters, etc.), dynamically adjust model parameters based on new data to improve prediction accuracy; for example, if the actual travel time of a transport route is 10 minutes longer than the time predicted by the simulation, this may be due to traffic congestion or other unforeseen factors; analyze the causes of these errors, such as whether they are caused by factors such as weather conditions, changes in traffic flow, etc.; if it is found that a specific section of road is often congested, the speed prediction model of that section can be automatically adjusted through an online learning algorithm to more accurately reflect the actual situation; continuously update and revise the comprehensive hydrogen consumption model based on real-time data so that it can more accurately reflect the current operating conditions; set up a periodic evaluation plan and regularly check the performance of the revised model to ensure that it is always in the optimal state; create a closed-loop control system to compare actual operating data with predicted values, identify sources of errors, and optimize the model accordingly; for example, if it is found that hydrogen consumption is higher than expected over a period of time, relevant parameters in the model, such as rolling resistance coefficient or wind resistance coefficient, can be adjusted through real-time data; conduct a comprehensive evaluation of the model's performance every quarter to ensure that it still meets the latest operational requirements.If the model prediction error is found to increase, the latest data will be used for retraining immediately; assuming that abnormal traffic conditions are encountered during a certain transportation process, resulting in frequent vehicle starts and stops, the system will promptly update the model parameters to adapt to this new operating condition; based on the revised comprehensive hydrogen consumption model, the simulation calculation is re-performed to generate new simulation results; the new simulation results are analyzed in depth to evaluate the selection of different transportation routes and hydrogen consumption estimates to ensure that they are always maintained at the optimal level; with the introduction of new data, the model and simulation calculations are continuously optimized to ensure the long-term reliability and continuous improvement capabilities of the system; for example, if it is found that the traffic conditions of a certain route have changed significantly within a specific time period, the simulation calculation can be re-performed to consider the impact of these changes; by analyzing the new simulation results, the traffic efficiency and hydrogen consumption of each route are evaluated, and the most appropriate transportation route is selected; assuming that certain influencing factors (such as road construction, policy changes, etc.) change over time, the system can continuously optimize the model and simulation calculations based on the latest data and conditions.

[0132] The effects of the above technical solutions are as follows: establish a real-time data acquisition system to ensure the accuracy and integrity of key operating data, and use encryption technology and security protocols to ensure the security of data during transmission and storage; enhance data security, prevent sensitive information from being leaked or tampered with, and improve system reliability and user trust; compare actual operating data with simulation results, evaluate model accuracy, identify error sources and analyze causes, and use online learning algorithms to dynamically correct models; improve the prediction accuracy of the model, reduce prediction errors, and ensure the accuracy of decision-making basis through continuous comparison and adjustment; dynamically correct the comprehensive hydrogen consumption model based on real-time data, and regularly evaluate the performance of the corrected model to ensure that the model is always in the best condition. The model is kept in an optimal state; the adaptability and long-term reliability of the model are maintained, so that it can respond to the actual changes in a timely manner, always maintain the optimal state, and avoid prediction deviations caused by model aging; simulation calculations and optimization analysis are re-performed based on the revised model to ensure that the selection of transportation routes and hydrogen consumption estimates are always maintained at the optimal level; the transportation route selection is optimized, hydrogen consumption is reduced, operational efficiency is improved, unnecessary fuel waste is reduced, and operating costs are directly reduced; more accurate hydrogen consumption estimates and optimized management help reduce energy consumption, indirectly reduce operating costs, and improve the company's profitability; by optimizing transportation routes and hydrogen consumption estimates, fuel consumption is reduced, operating costs are further reduced, and economic benefits are improved.

[0133] One embodiment of the present invention, as Figure 2 As shown, a vehicle transportation route hydrogen consumption simulation system based on multiple models, the system comprises:

[0134] Data acquisition module: Based on the existing basic power demand model and hydrogen consumption calculation model, multi-source data is collected in real time through a multi-source data acquisition network; the multi-source data is combined with historical data, the existing model is optimized through a machine learning algorithm, and the hydrogen consumption calculation model is updated;

[0135] Model building module: optimizes the existing environmental impact model through new data of external environmental factors collected in real time, and further corrects the hydrogen consumption calculation results according to the impact of environmental conditions on hydrogen fuel cell efficiency;

[0136] Model integration module: Integrate the optimized basic power demand model, hydrogen consumption calculation model and environmental impact model to form a multi-level comprehensive hydrogen consumption model;

[0137] Result simulation module: input relevant data of different transportation routes, simulate and calculate different transportation routes based on the comprehensive hydrogen consumption model, and obtain the hydrogen consumption of each route under different working conditions; output the simulation results of hydrogen consumption of each transportation route under different working conditions, and continuously optimize the model by comparing with the actual operation data;

[0138] Route optimization module: Based on the latest simulation results, the optimal transportation route is determined through a multi-objective optimization algorithm; the optimal transportation route recommendation is output, and the transportation route is dynamically adjusted according to the actual operation situation;

[0139] Dynamic correction module: During the transportation process, actual operation data is collected in real time; based on the actual operation data, the comprehensive hydrogen consumption model is re-evaluated and optimized.

[0140] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A multi-model based vehicle transport route hydrogen consumption simulation method, characterized in that: The method comprises: S1. Based on the basic power demand model and the hydrogen consumption calculation model, multi-source data collected in real time through the multi-source data collection network is combined with the historical data, the existing model is optimized through the machine learning algorithm, and the hydrogen consumption calculation model is updated; S2. Optimize the environmental impact model through new data of external environmental factors collected in real time, and further correct the hydrogen consumption calculation results according to the impact of environmental conditions on hydrogen fuel cell efficiency; S3. Integrate the optimized basic power demand model, hydrogen consumption calculation model and environmental impact model to form a multi-level comprehensive hydrogen consumption model; S4. Input relevant data of different transportation routes, perform simulation calculations on different transportation routes based on the comprehensive hydrogen consumption model, and obtain the hydrogen consumption of each route under different working conditions; output the simulation results of hydrogen consumption of each transportation route under different working conditions, and continuously optimize the model by comparing the actual operation data; S5. Based on the simulation results, determine the best transportation route through a multi-objective optimization algorithm; output the optimized best transportation route suggestion, and dynamically adjust the transportation route according to the actual operation situation; S6. During the transportation process, actual operation data is collected in real time; based on the actual operation data, the comprehensive hydrogen consumption model is re-evaluated and optimized.

2. According to claim 1, a multi-model based vehicle transportation route hydrogen consumption simulation method is characterized in that: Said S1 comprises: S11. Define the type and location of data collection points, and design data collection frequency and accuracy requirements; S12. Build a data transmission network, encrypt the data transmission process based on the data encryption algorithm, integrate the data collection system, and uniformly receive, store and manage multi-source data; S13, obtaining historical data, and preprocessing the collected real-time data and historical data; S14. Based on the vehicle motion equation model and vehicle dynamics principles, combined with newly collected real-time vehicle operation data, optimize the calculation of basic power requirements under different working conditions; S15. Based on the power system response characteristic curve and efficiency curve model, the new hydrogen fuel cell working characteristic data is used to optimize the relationship between hydrogen consumption and electric energy output, and a new hydrogen consumption calculation model is established; S16. Combine the existing basic power demand model with the hydrogen consumption calculation model, further optimize it through new data, and calculate the hydrogen consumption under different operating conditions.

3. According to claim 2, a multi-model based vehicle transportation route hydrogen consumption simulation method is characterized in that: The S14 comprises: S141. Using the existing mass change measurement model, add new data to optimize the dynamic mass model, and optimize the drag coefficient by combining the latest wind speed and wind direction data through CFD simulation; S142, based on the rolling resistance coefficient model, using new data to perform dynamic adjustment; the new data includes tire wear degree, tire type, tire pressure and road type; S143. Combine the new data from the high-precision slope sensor to optimize the slope and climbing resistance relationship model, and combine Newton's second law and vehicle dynamics parameters to construct and optimize the dynamic equation of vehicle motion; S144, dynamically adjust the parameters in the dynamic equation according to the new data collected in real time to optimize the power demand prediction; S145. Analyze new and old data using machine learning or pattern recognition algorithms to automatically identify driving conditions; S146. Optimize the power demand prediction model based on the operating condition identification result.

4. The method for simulating hydrogen consumption of a vehicle transportation route based on multiple models according to claim 2 is characterized in that: The S15 comprises: S151. Optimize the response characteristic curve and efficiency curve of the power system based on the measured data through simulation testing; S152, evaluate the adaptability of transient working conditions, combine the chemical reaction principle of hydrogen fuel cells, and optimize the efficiency model; S153. Analyze temperature variation patterns, optimize temperature compensation mechanism and aging prediction model, and dynamically adjust hydrogen consumption calculation parameters; S154, optimizing and matching basic power demand and power system output capacity according to real-time operation data and power system response characteristics; S155. Establish a real-time feedback mechanism to dynamically adjust the correction strategy based on the actual operation conditions and hydrogen consumption calculation results, and continuously optimize the hydrogen consumption calculation model framework; S156. Integrate the optimized power system response characteristic curve and efficiency curve as well as the hydrogen consumption calculation model framework into the hydrogen consumption calculation model, verify the model accuracy, and continuously optimize based on feedback.

5. The method for simulating hydrogen consumption of a vehicle transportation route based on multiple models according to claim 1, characterized in that: The S2 comprises: S21. Identify environmental factors that affect hydrogen fuel cell efficiency and classify these factors based on models and new data; S22. Evaluate the importance and relevance of each environmental factor, determine the factors that should be considered in the model, and update the evaluation results using new data; S23. Based on the existing test methods, test the hydrogen fuel cell under different environments and use the new data to supplement or correct the performance parameter records; S24. Based on the new test data, the mapping relationship between environmental factors and hydrogen fuel cell efficiency is updated using statistical analysis methods; S25. Evaluate the fit and predictive ability of the preliminary model, introduce complexity assessment indicators, determine whether the model needs to be simplified or complicated, and verify it with new data; S26. Identify deficiencies in the model and design supplementary tests or collect more relevant data to enrich the training set; S27. Apply feature engineering techniques to optimize the input feature set, and modify and retrain the model through machine learning algorithms; S28, introduce judgment logic to guide model selection and optimization direction, set conditional branches to handle model applications under different environmental conditions; S29. Use adaptive mechanisms to enable the model to automatically adjust parameters or select the best version, regularly evaluate model performance, and trigger the update process when performance degradation is found, using the latest environment and performance data for retraining and verification.

6. The method for simulating hydrogen consumption of a vehicle transportation route based on multiple models according to claim 1, characterized in that: The S3 includes: S31. Layer the factors affecting hydrogen consumption and conduct an in-depth analysis of the factors affecting each layer based on the existing model; S32. For each level of influencing factors, construct or optimize a sub-model based on the existing sub-model and in combination with new data, and integrate the sub-models through a Bayesian network; S33. Conduct preliminary analysis and evaluation of the comprehensive hydrogen consumption model, use historical data to backtest the effectiveness of the model, and dynamically adjust the model based on new data; S34, during the model prediction process, automatically switching to the sub-model that best suits the current situation through conditional judgment logic; S35. Dynamically adjust the model parameters or structure according to the real-time data and prediction results, establish a feedback mechanism to compare the actual hydrogen consumption data with the predicted value, identify the source of error, and optimize the model accordingly; S36. Use particle swarm optimization and other algorithms to further optimize the model, introduce new influencing factors and characteristic variables to improve the model structure; S37. Use an independent test set to validate the optimized model, and regularly update and improve the model.

7. The method for simulating hydrogen consumption of a vehicle transportation route based on multiple models according to claim 1, characterized in that: The S4 comprises: S41. Plan multiple transport routes, consider the impact of traffic conditions at different time periods, and optimize them by using existing traffic prediction models and adding new data; S42, based on historical meteorological data and real-time weather forecast, simulate the transportation environment through simulation software; S43, setting vehicle parameters for a specific vehicle model, inputting corresponding parameters of the hydrogen fuel cell, and optimizing using the existing vehicle dynamics model combined with the new data; S44, simulating different time and route conditions based on real-time traffic data and historical statistical data, and using cloud computing resources to perform simulation calculations of multiple routes in parallel; S45, recording key data in the simulation process in real time, outputting simulation results, and using the simulation results for subsequent model optimization and verification.

8. The method for simulating hydrogen consumption of a vehicle transportation route based on multiple models according to claim 1, characterized in that: The S5 comprises: S51. Clarify the optimization goals, set weights based on the enterprise operation strategy and market demand, and build a multi-objective optimization function; S52, evaluate the applicability of different multi-objective optimization algorithms and adjust algorithm parameters based on new data; S53. Use the simulation results as input to run the multi-objective optimization algorithm, generate the Pareto frontier solution set, analyze the performance of different solutions, and select the best transportation route from the solution set, while using new data to continuously optimize the selection process.

9. The method for simulating hydrogen consumption of a vehicle transportation route based on multiple models according to claim 1, characterized in that: The S6 comprises: S61. Establish a real-time data collection system to continuously monitor key data during vehicle operation and use encryption technology and security protocols to ensure data security; S62. Compare the actual operation data with the simulation results, evaluate the model accuracy, identify the source of error and analyze the cause, and dynamically correct the model using the online learning algorithm; S63. Dynamically revise the comprehensive hydrogen consumption model based on real-time data and regularly evaluate the performance of the revised model; S64. Re-perform simulation calculation and optimization analysis based on the revised model.

10. A vehicle transportation route hydrogen consumption simulation system based on multiple models, characterized in that: The system comprises: Data acquisition module: Based on the basic power demand model and hydrogen consumption calculation model, multi-source data is collected in real time through a multi-source data acquisition network; the multi-source data is combined with historical data, the existing model is optimized through a machine learning algorithm, and the hydrogen consumption calculation model is updated; Model building module: optimizes the environmental impact model through new data of external environmental factors collected in real time, and further corrects the hydrogen consumption calculation results according to the impact of environmental conditions on hydrogen fuel cell efficiency; Model integration module: Integrate the optimized basic power demand model, hydrogen consumption calculation model and environmental impact model to form a multi-level comprehensive hydrogen consumption model; Result simulation module: input relevant data of different transportation routes, simulate and calculate different transportation routes based on the comprehensive hydrogen consumption model, and obtain the hydrogen consumption of each route under different working conditions; output the simulation results of hydrogen consumption of each transportation route under different working conditions, and continuously optimize the model by comparing with the actual operation data; Route optimization module: Based on the latest simulation results, the optimal transportation route is determined through a multi-objective optimization algorithm; the optimal transportation route recommendation is output, and the transportation route is dynamically adjusted according to the actual operation situation; Dynamic correction module: During the transportation process, actual operation data is collected in real time; based on the actual operation data, the comprehensive hydrogen consumption model is re-evaluated and optimized.

Images