Efficient power and energy management system of emergency energy vehicle

Abstract

The invention relates to the technical field of energy management, in particular to an emergency energy vehicle efficient power and energy management system which comprises a methanol reforming device, a hydrogen storage device, a fuel cell and an energy management module. The hydrogen storage device is used for ensuring stable supply of hydrogen, the fuel cell converts the hydrogen into electric energy to drive the vehicle, redundant electric energy is stored in the battery pack, and the energy management module optimizes energy distribution and supply according to the running state and energy requirements of the vehicle. Through integration of the multiple energy supply units, dynamic adjustment of the intelligent management module, energy prediction and optimization and use of the regenerative braking system, the energy efficiency and reliability of the vehicle are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management, and in particular to a high-efficiency power and energy management system for an emergency energy vehicle. Background Art

[0002] Energy Management System (EMS) plays a vital role in improving vehicle energy efficiency and reliability. Especially in emergency energy vehicles, efficient energy management can significantly improve vehicle performance and sustainability.

[0003] There are currently the following deficiencies:

[0004] 1) Traditional emergency energy vehicles usually rely on a single energy supply method, such as pure batteries or pure fuel engines. This method may not be able to meet the energy needs of the vehicle under different working conditions in some cases, especially under high load or acceleration.

[0005] 2) The energy distribution in traditional systems is usually static, that is, the contribution ratio of each energy supply unit is pre-set. This static distribution method cannot flexibly respond to changes in vehicle conditions, which may lead to energy waste or insufficient energy supply.

[0006] 3) Traditional systems lack the ability to predict future energy demand. This results in the system being unable to adjust energy supply in a timely manner when faced with emergencies, thus affecting vehicle performance and efficiency.

[0007] 4) Energy management in traditional systems often requires human intervention to manually adjust the working status of each energy supply unit. This approach is not only inefficient but also prone to errors. Summary of the invention

[0008] The present invention provides an efficient power and energy management system for emergency energy vehicles, which significantly improves the energy efficiency and reliability of the vehicle through the integration of multiple energy supply units, dynamic adjustment of intelligent management modules, energy prediction and optimization, and the use of a regenerative braking system.

[0009] In order to achieve the purpose of the present invention, the technical solution adopted is: an efficient power and energy management system for emergency energy vehicles, including a methanol reforming device, a hydrogen storage device, a fuel cell and an energy management module. The methanol reforming device converts methanol into hydrogen and stores it in the hydrogen storage device. The hydrogen storage device is used to ensure a stable supply of hydrogen. The fuel cell converts hydrogen into electricity to drive the vehicle, and the excess electricity is stored in the battery pack. The energy management module optimizes energy distribution and supply according to the vehicle's operating status and energy requirements.

[0010] As an optimization scheme of the present invention, the energy management module obtains the vehicle's speed, acceleration and load information through the on-board sensors to determine the vehicle's operating status; when the vehicle accelerates, the energy management module increases the hydrogen production rate of the methanol reforming device to ensure an adequate supply of hydrogen; according to the acceleration demand, the output power of the fuel cell is adjusted to meet the instantaneous increase in energy demand; when there is a high acceleration demand, the battery pack is enabled to assist in power supply to smooth out fluctuations in energy demand and avoid overloading the fuel cell.

[0011] As an optimization solution of the present invention, when the vehicle decelerates or brakes, energy is recovered through the regenerative braking system and stored in the battery pack; when the vehicle accelerates or is highly loaded, the recovered electric energy is used preferentially.

[0012] As an optimization scheme of the present invention, a machine learning algorithm is used to predict future energy demand, and the hydrogen production rate of the methanol reforming device and the output power of the fuel cell are adjusted in advance, specifically:

[0013] 1) Collect historical operating data of the vehicle, including the hydrogen production rate of the methanol reformer, the output power of the fuel cell, and the total load of the vehicle;

[0014] 2) Data preprocessing: normalize the data to the same scale and extract features including velocity change rate, acceleration change rate, and load change rate;

[0015] 3) Divide the data into training set, validation set and test set, select the random forest algorithm, use the training set data to train the random forest model, use the validation set data to adjust the parameters, and use the test set to evaluate the random forest model;

[0016] 4) Use the trained random forest model to predict future energy demand based on real-time data, and adjust the hydrogen production rate of the methanol reforming unit and the output power of the fuel cell in advance based on the predicted energy demand.

[0017] As an optimization solution of the present invention, a dynamic energy allocation strategy is used to optimize energy distribution and supply, specifically:

[0018] A. Define the key parameters of the vehicle in operation;

[0019] B. Define executable actions to adjust the contribution ratio of each energy supply unit of the methanol reformer, fuel cell and battery pack;

[0020] C. Design a reward function to give rewards or penalties based on the degree of satisfaction of energy requirements, system efficiency, and system stability;

[0021] D. Use reinforcement learning algorithm to train the model and find the optimal energy allocation solution through trial and error learning.

[0022] As an optimization scheme of the present invention, in step B, executable actions include: increasing the hydrogen production rate of the methanol reforming device, reducing the hydrogen production rate of the methanol reforming device, increasing the output power of the fuel cell, reducing the output power of the fuel cell, enabling battery pack auxiliary power supply, stopping battery pack auxiliary power supply and adjusting the recovery efficiency of the regenerative braking system.

[0023] As an optimization solution of the present invention, in step D, it specifically includes:

[0024] a. Build a simulation environment to simulate the running state of the vehicle;

[0025] b. Select the PPO reinforcement learning algorithm and initialize the model parameters;

[0026] c. Collect vehicle operation data in a simulated environment, including actions, status, and rewards;

[0027] d. Use the collected data to train the model and find the optimal energy allocation strategy through trial and error learning.

[0028] The present invention has positive effects: 1) The present invention ensures that each energy supply unit works in an optimal state by dynamically adjusting the energy supply, thereby reducing energy waste; and utilizes the regenerative braking system to recover energy, thereby further improving energy efficiency;

[0029] 2) The redundant design of the multi-energy supply unit of the present invention ensures that when a unit fails, other units can continue to provide energy, thereby improving the reliability of the system;

[0030] 3) The dynamic energy allocation strategy of the present invention ensures that the vehicle can operate smoothly under different working conditions such as acceleration, deceleration and high load. The efficient energy management reduces the consumption of fuel and electricity and reduces the overall cost;

[0031] 4) The reinforcement learning algorithm of the present invention can adaptively adjust the energy supply strategy to adapt to different driving environments and working conditions, thereby improving the adaptability and flexibility of the system; through machine learning and reinforcement learning algorithms, intelligent energy management and prediction are achieved, thereby improving the intelligence level and management efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0033] Figure 1 It is a principle block diagram of the present invention;

[0034] Among them: 1. Methanol reforming unit, 2. Hydrogen storage unit, 3. Fuel cell, 4. Energy management module, 5. Battery pack. DETAILED DESCRIPTION

[0035] like Figure 1 As shown, the present invention discloses an efficient power and energy management system for an emergency energy vehicle, comprising a methanol reforming device 1, a hydrogen storage device 2, a fuel cell 3 and an energy management module 4. The methanol reforming device 1 converts methanol into hydrogen and stores it in the hydrogen storage device 2. The hydrogen storage device 2 is used to ensure a stable supply of hydrogen. The fuel cell 3 converts hydrogen into electric energy to drive the vehicle, and the excess electric energy is stored in a battery pack 5. The energy management module 4 optimizes energy distribution and supply according to the operating status and energy demand of the vehicle.

[0036] The methanol reformer 1 converts methanol (CH3OH) into hydrogen (H2) and other byproducts through a chemical reaction, and decomposes methanol into hydrogen and carbon dioxide through a high-temperature catalytic reaction. The generated hydrogen is transported to the hydrogen storage device 2 to ensure a continuous supply of hydrogen. The hydrogen storage device 2 is used to store the hydrogen generated by the methanol reformer 1 to ensure a stable supply of hydrogen. The fuel cell 3 converts hydrogen and oxygen into electrical energy through an electrochemical reaction to drive the vehicle's electric motor. The electrochemical reaction efficiency of the fuel cell is high, and the energy conversion efficiency can reach more than 60%, which is much higher than that of a traditional internal combustion engine. The only emission of the fuel cell is water, and there is no harmful gas emission, which is very environmentally friendly.

[0037] The battery pack 5 is used to store excess electrical energy, which can come from the excess output of the fuel cell or the energy recovered by the regenerative braking system. When the vehicle is accelerating or under high load, the battery pack 5 can provide additional electrical energy to smooth fluctuations in energy demand and avoid overloading the fuel cell 3. When the vehicle is decelerating or braking, energy is recovered by the regenerative braking system and stored in the battery pack 5 to improve overall energy efficiency. The battery pack can quickly respond to the instantaneous energy needs of the vehicle and improve the driving experience.

[0038] The energy management module 4 monitors the vehicle's speed, acceleration, load and other information in real time through the vehicle's onboard sensors to determine the vehicle's operating status.

[0039] Dynamic adjustment: Dynamically adjust the hydrogen production rate of the methanol reformer, the output power of the fuel cell and the auxiliary power supply of the battery pack according to the real-time status of the vehicle and the predicted energy demand.

[0040] Energy optimization: By optimizing energy distribution and supply, it ensures that the vehicle can operate efficiently under different working conditions, improving energy efficiency and reliability.

[0041] Prediction and control: Use machine learning algorithms to predict future energy demand and adjust the hydrogen production rate of the methanol reformer and the output power of the fuel cell in advance to avoid system overload or energy shortage.

[0042] Energy management module 4 can make intelligent decisions and optimize energy distribution based on real-time data and prediction models. Through dynamic adjustment and optimization, it ensures that the system can operate stably under various working conditions and improves reliability. Energy management module 4 can adaptively adjust energy supply strategies to adapt to different driving environments and working conditions.

[0043] Through the integration of multiple energy supply units and dynamic adjustment of intelligent management modules, the vehicle can be ensured to operate efficiently under different working conditions and improve overall energy efficiency.

[0044] The redundant design of the multi-energy supply unit and the dynamic adjustment of the intelligent management module ensure that the system can operate stably under various working conditions and improve reliability.

[0045] The use of the methanol reforming device 1 and the fuel cell 3 reduces the pollution of traditional fuel engines and meets environmental protection requirements.

[0046] Use machine learning algorithms to predict future energy demand and adjust energy supply in advance to avoid system overload or energy shortage. Energy management module 4 can adaptively adjust energy supply strategy to adapt to different driving environments and working conditions to improve driving experience.

[0047] The energy management module 4 obtains the speed, acceleration and load information of the vehicle through the on-board sensors to determine the running status of the vehicle; when the vehicle accelerates, the energy management module 4 increases the hydrogen production rate of the methanol reforming device 1 to ensure sufficient supply of hydrogen; according to the acceleration demand, the output power of the fuel cell 3 is adjusted to meet the instantaneous increase in energy demand; when there is a high acceleration demand, the battery pack 5 is enabled to assist in power supply to smooth the fluctuation of energy demand and avoid overloading of the fuel cell 3. Among them, the on-board sensors include GPS speed sensors to measure the driving speed of the vehicle in real time and provide accurate speed data. The three-axis acceleration sensor measures the acceleration of the vehicle in real time and provides accurate acceleration data. The Hall effect current sensor measures the current of the vehicle electrical system in real time to help the energy management module monitor the use of electric energy. The voltage transformer measures the voltage of the vehicle electrical system in real time to help the energy management module monitor the use of electric energy. The pressure sensor measures the pressure of the vehicle tires, suspension system and other parts in real time to help the energy management module determine the load and driving status of the vehicle.

[0048] When the vehicle is decelerating or braking, energy is recovered through the regenerative braking system and stored in the battery pack 5; when the vehicle is accelerating or under high load, the recovered electric energy is used preferentially. Energy recovery during deceleration or braking; when the vehicle is decelerating or braking, the motor works in reverse and becomes a generator, converting the kinetic energy of the vehicle into electric energy. The converted electric energy is stored in the battery pack 5 through the inverter and the charging circuit. By recovering the kinetic energy during vehicle deceleration or braking, energy waste is reduced and overall energy efficiency is improved. The recovered energy can be used for subsequent acceleration or high load requirements to extend the vehicle's cruising range. The regenerative braking system can reduce the burden of the traditional friction braking system, reduce the wear of the brake pads, and extend the service life of the braking system. When the vehicle is accelerating or under high load, the battery pack 5 can provide additional electric energy to smooth fluctuations in energy demand and avoid overloading the fuel cell 3. The battery pack 3 can quickly respond to the instantaneous energy demand of the vehicle and improve the driving experience. The battery pack 3 can work in conjunction with other energy supply units (such as fuel cells) to provide a flexible energy method.

[0049] The machine learning algorithm is used to predict future energy demand and adjust the hydrogen production rate of the methanol reforming device 1 and the output power of the fuel cell 3 in advance, specifically:

[0050] 1) Collect historical operation data of the vehicle, including the hydrogen production rate of the methanol reformer 1, the output power of the fuel cell 3 and the total load of the vehicle; the total load of the vehicle includes passenger weight, cargo weight, vehicle speed, acceleration, etc. In order to collect these data, it is necessary to install corresponding sensors on the vehicle, install a hydrogen production rate sensor and a temperature sensor on the methanol reformer 1. Install an output power sensor and a temperature sensor on the fuel cell 3. Install a speed sensor, an acceleration sensor, a load sensor, a current sensor and a voltage sensor on the vehicle.

[0051] 2) Data preprocessing: normalize the data to the same scale and extract features including velocity change rate, acceleration change rate, and load change rate;

[0052] Data cleaning is performed during data preprocessing: outliers and noise are removed to ensure data accuracy. Data is normalized to the same scale to facilitate subsequent machine learning processing. Useful features are extracted, such as the rate of change of speed, rate of change of acceleration, and rate of change of load.

[0053] 3) Divide the data into training set, validation set and test set, select the random forest algorithm, use the training set data to train the random forest model, use the validation set data to adjust the parameters, and use the test set to evaluate the random forest model;

[0054] During implementation, read the CSV file vehicle_data.csv containing historical operation data, including timestamp, hydrogen production rate, fuel cell output power, vehicle speed, vehicle acceleration, vehicle load, current and voltage. Divide the data into training set, validation set and test set. The usual ratio is 70% training set, 15% validation set and 15% test set. Use the training set data to train the random forest model, use the validation set data for hyperparameter tuning, and use the test set data to evaluate the performance of the model.

[0055] 4) Use the trained random forest model to predict future energy demand based on real-time data, and adjust the hydrogen production rate of the methanol reformer 1 and the output power of the fuel cell 3 in advance based on the predicted energy demand. This helps optimize energy distribution and supply and improve the energy efficiency and reliability of the vehicle. The continuous monitoring and adjustment mechanism ensures that the system can always maintain the best state under different working conditions.

[0056] The energy management module 4 uses a dynamic energy allocation strategy to optimize energy distribution and supply according to the vehicle's operating status and energy demand, specifically:

[0057] A. Define the key parameters of the vehicle in operation. These parameters are used to describe the current state of the vehicle in order to dynamically adjust energy distribution. The key parameters include the current vehicle speed, acceleration, load, battery status, road slope and weather conditions.

[0058] B. defining executable actions to adjust the contribution ratio of each energy supply unit of the methanol reforming device 1, the fuel cell 3 and the battery group 5;

[0059] In step B, the executable actions include: increasing the hydrogen production rate of the methanol reformer, reducing the hydrogen production rate of the methanol reformer, increasing the output power of the fuel cell, reducing the output power of the fuel cell, enabling the battery pack to provide auxiliary power, stopping the battery pack to provide auxiliary power, and adjusting the recovery efficiency of the regenerative braking system.

[0060] C. Design a reward function to give rewards or penalties based on the degree of satisfaction of energy requirements, system efficiency, and system stability;

[0061] Energy requirement satisfaction: rewards or penalties are given based on the gap between predicted energy requirements and actual energy provided.

[0062] System efficiency: rewards are given based on the overall energy efficiency of the system (such as energy conversion efficiency, energy utilization rate).

[0063] System stability: rewards or penalties are given based on indicators such as the system's stable operating time and failure rate.

[0064] D. Use the reinforcement learning algorithm to train the model and find the optimal energy allocation solution through trial and error learning. In step D, it specifically includes:

[0065] a. Build a simulation environment to simulate the vehicle's running state; initialize the environment state. Select an action based on the current state. After executing the action, observe the new state and reward. Store the state, action, reward, and new state in the experience playback buffer.

[0066] b. Select the PPO reinforcement learning algorithm and initialize the model parameters; define the PPO model, including the policy network (Actor) and the value network (Critic).

[0067] c. Collect vehicle operation data in a simulated environment, including actions, status, and rewards;

[0068] d. Use the collected data to train the model and find the optimal energy allocation strategy through trial and error learning. During implementation, calculate the return (Return) of each time step, and then use the PPO algorithm to update the model parameters. During the update, calculate the advantage function, calculate the probability of the old strategy, perform multiple gradient descents, calculate the probability ratio, and calculate the loss function to update the model parameters.

[0069] Return:

[0070]

[0071] Where: R t is the return, γ is the discount factor, r t+k+1 is the reward at step t+k+1.

[0072] Advantage function:

[0073] A t =R t –V(s t )

[0074] Where: V(s t ) is the state s t The value function of .

[0075] Probability ratio:

[0076]

[0077] Where: π θ (a t ∣s t ) is the action a taken under the new strategy t The probability of θold (a t ∣s t ) is the old strategy to take action at probability.

[0078] The beneficial effects of the present invention are specifically embodied in:

[0079] 1) Efficient energy conversion and utilization

[0080] Methanol reforming unit: Efficiently converts methanol into hydrogen through high-temperature catalytic reactions to ensure a continuous supply of hydrogen.

[0081] Fuel cell: converts hydrogen and oxygen into electrical energy through electrochemical reactions, with an energy conversion efficiency of over 60%, much higher than traditional internal combustion engines.

[0082] Battery Pack: Stores excess electrical energy, including excess output from the fuel cell and energy recovered by the regenerative braking system, improving overall energy efficiency.

[0083] Zero emission: The only emission of fuel cells is water, no harmful gas emissions, very environmentally friendly. Reduced pollution: Compared with traditional fuel engines, the use of methanol reforming devices and fuel cells significantly reduces environmental pollution.

[0084] Dynamic energy management: The vehicle's speed, acceleration, load and other information are monitored in real time through on-board sensors to determine the vehicle's operating status. Based on the vehicle's real-time status and predicted energy demand, the hydrogen production rate of the methanol reforming unit, the output power of the fuel cell and the auxiliary power supply of the battery pack are dynamically adjusted. By optimizing energy distribution and supply, the vehicle can be ensured to operate efficiently under different operating conditions, improving energy efficiency and reliability.

[0085] Machine learning: Use machine learning algorithms to predict future energy demand and adjust the hydrogen production rate of the methanol reformer and the output power of the fuel cell in advance to avoid system overload or energy shortage. The energy management module can adaptively adjust the energy supply strategy to adapt to different driving environments and working conditions to improve the driving experience.

[0086] Energy recovery: When the vehicle decelerates or brakes, energy is recovered through the regenerative braking system and stored in the battery pack, reducing energy waste and improving overall energy efficiency.

[0087] Extended driving range: The recovered energy can be used for subsequent acceleration or high load demands, extending the vehicle's driving range.

[0088] Reduce wear: Reduce the burden on traditional friction brake systems, reduce brake pad wear, and extend the service life of the brake system.

[0089] Redundant design: The integration of multiple energy supply units and the dynamic adjustment of intelligent management modules ensure that the system can operate stably under various working conditions and improve reliability.

[0090] Flexible energy mode: Battery packs can work together with other energy supply units (such as fuel cells) to provide flexible energy mode.

[0091] Key parameters: Defines the key parameters of the vehicle in operation, including speed, acceleration, load, battery status, road slope and weather conditions, which are used to dynamically adjust energy distribution.

[0092] Executable actions: Executable actions are defined, including increasing or decreasing the hydrogen production rate of the methanol reformer, increasing or decreasing the output power of the fuel cell, enabling or disabling the auxiliary power supply of the battery pack, and adjusting the recovery efficiency of the regenerative braking system.

[0093] PPO algorithm: Use the PPO (Proximal Policy Optimization) reinforcement learning algorithm to find the optimal energy allocation strategy through trial and error learning.

[0094] Reward function: A reward function is designed to give rewards or penalties based on the degree of satisfaction of energy requirements, system efficiency, and system stability to ensure that the model is continuously optimized during training.

[0095] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The high-efficiency power and energy management system for emergency energy vehicles is characterized by: The invention comprises a methanol reforming device (1), a hydrogen storage device (2), a fuel cell (3) and an energy management module (4); the methanol reforming device (1) converts methanol into hydrogen and stores it in the hydrogen storage device (2); the hydrogen storage device (2) is used to ensure a stable supply of hydrogen; the fuel cell (3) converts hydrogen into electric energy to drive the vehicle, and the excess electric energy is stored in a battery pack (5); and the energy management module (4) optimizes energy distribution and supply according to the operating state and energy demand of the vehicle.

2. The high-efficiency power and energy management system for emergency energy vehicles according to claim 1 is characterized in that: The energy management module (4) obtains the speed, acceleration and load information of the vehicle through the on-board sensor to judge the running state of the vehicle; when the vehicle accelerates, the energy management module (4) increases the hydrogen production rate of the methanol reforming device (1) to ensure sufficient supply of hydrogen; according to the acceleration demand, the output power of the fuel cell (3) is adjusted to meet the instantaneous increase in energy demand; when the acceleration demand is high, the battery group (5) is enabled to assist in power supply to smooth the fluctuation of energy demand and avoid overloading of the fuel cell (3).

3. The high-efficiency power and energy management system for emergency energy vehicles according to claim 2 is characterized in that: When the vehicle is decelerating or braking, energy is recovered through the regenerative braking system and stored in the battery pack (5); when the vehicle is accelerating or under high load, the recovered electric energy is used preferentially.

4. The high-efficiency power and energy management system for emergency energy vehicles according to claim 3 is characterized in that: The machine learning algorithm is used to predict future energy demand, and the hydrogen production rate of the methanol reforming device (1) and the output power of the fuel cell (3) are adjusted in advance, specifically: 1) collecting historical operation data of the vehicle, including the hydrogen production rate of the methanol reforming device (1), the output power of the fuel cell (3) and the total load of the vehicle; 2) Data preprocessing: normalize the data to the same scale and extract features including velocity change rate, acceleration change rate, and load change rate; 3) Divide the data into training set, validation set and test set, select the random forest algorithm, use the training set data to train the random forest model, use the validation set data to adjust the parameters, and use the test set to evaluate the random forest model; 4) Using the trained random forest model, the future energy demand is predicted based on the real-time data, and the hydrogen production rate of the methanol reforming device (1) and the output power of the fuel cell (3) are adjusted in advance based on the predicted energy demand.

5. The high-efficiency power and energy management system for emergency energy vehicles according to claim 4 is characterized in that: Use dynamic energy allocation strategies to optimize energy distribution and supply, specifically: A. Define the key parameters of the vehicle in operation; B. defining executable actions to adjust the contribution ratio of each energy supply unit of the methanol reformer (1), the fuel cell (3) and the battery pack (5); C. Design a reward function to give rewards or penalties based on the degree of satisfaction of energy requirements, system efficiency, and system stability; D. Use reinforcement learning algorithm to train the model and find the optimal energy allocation solution through trial and error learning.

6. The high-efficiency power and energy management system for emergency energy vehicles according to claim 5 is characterized in that: In step B, executable actions include: increasing the hydrogen production rate of the methanol reformer, reducing the hydrogen production rate of the methanol reformer, increasing the output power of the fuel cell, reducing the output power of the fuel cell, enabling battery pack auxiliary power supply, stopping battery pack auxiliary power supply and adjusting the recovery efficiency of the regenerative braking system.

7. The high-efficiency power and energy management system for emergency energy vehicles according to claim 5 is characterized in that: In step D, it specifically includes: a. Build a simulation environment to simulate the running state of the vehicle; b. Select the PPO reinforcement learning algorithm and initialize the model parameters; c. Collect vehicle operation data in a simulated environment, including actions, status, and rewards; d. Use the collected data to train the model and find the optimal energy allocation strategy through trial and error learning.

Images