HESS energy management control method and device, vehicle and medium

Abstract

The invention discloses an energy management control method and device, a vehicle and an HESS energy management control method. The HESS energy management control method comprises the steps that a vehicle end uploads HESS state parameters of the vehicle to a cloud end; the cloud dynamically plans the future driving condition of the vehicle by using traffic information collected in real time to obtain a speed-time curve and a slope-time curve; setting a time value, and extracting the future driving working conditions of the vehicle in the planned future driving working conditions of the vehicle within a time period of the previous set time value as available working condition information fragments; the cloud end obtains an optimal battery SOC control sequence in the available working condition information fragment according to the HESS state parameters of the vehicle and the available working condition information fragment; and the vehicle end performs energy management on the HESS according to the obtained optimal battery SOC control sequence. According to the invention, the energy management problem of the HESS in a complex and changeable environment is effectively solved, the energy utilization efficiency and economical efficiency of the vehicle are improved, and the service life of key parts is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and in particular to a HESS energy management control method. Background Art

[0002] Developing new energy public transportation is an important measure for my country to build green and smart cities. As an important part of the urban public transportation system, pure electric buses have advantages such as zero emissions and low noise, but they also have development bottlenecks such as short driving range and long charging time. Improving the economy and practicality of pure electric buses is of great significance to breaking through the bottleneck of their promotion and application. Single battery energy storage is the earliest energy storage solution applied to pure electric buses. The technology is relatively mature and has been widely used, but it has disadvantages such as low power density and limited cycle life. Compared with the traditional single battery energy storage solution, the hybrid energy storage system (HESS) composed of batteries and supercapacitors can have the advantages of high energy density and high power density, and improve the economy under different operating conditions by reasonably allocating energy flow.

[0003] HESS has a variety of different working modes, which can generally be divided into battery-dominated mode, supercapacitor-dominated mode and collaborative working mode. How to make working mode decisions and optimize energy allocation strategies is the key to improving the energy efficiency of hybrid energy storage systems under different operating conditions. At present, there have been relevant studies in this field, which can be summarized into: 1) rule-based methods and 2) optimization-based methods according to different control strategies.

[0004] The rule-based HESS control strategy is a strategy that uses preset rules to allocate power to batteries and supercapacitors based on the real-time vehicle power demand. This type of method sets a series of sophisticated logical rules based on expert experience and prior knowledge to achieve real-time control and optimization of the hybrid energy storage system. According to the nature of the rules, it can be further divided into two types: deterministic rule-based and fuzzy rule-based.

[0005] In deterministic rule-based strategies, an "if-then-else" model is usually adopted to set precise threshold values ​​and logic rules based on system state variables such as battery state of charge (SOC), required power, vehicle speed, etc.

[0006] For example, ZHAO et al. (ZHAO Y, WANG W, XIANG C, et al. Research and bench test of nonlinear model predictive control-based power allocation strategy for hybrid energy storage system. IEEE Access, 2018, 6(1): 70770–70787) strictly set the battery SOC range and required power conditions and used a logic threshold strategy to limit the battery power. When the SOC is lower than a certain threshold or the required power exceeds the preset range, the excess power is provided by the supercapacitor.

[0007] The strategy based on fuzzy rules introduces fuzzy logic control, maps input variables to fuzzy sets, and makes control decisions through fuzzy reasoning, thus achieving more flexible and intelligent control. For example, Zhou Meilan et al. (Zhou Meilan, Feng Jifeng, Zhang Yu, et al. Research on power distribution control strategy of hybrid energy storage system for pure electric buses. Transactions of China Electrotechnical Society, 2019, 34(23):5001–5013) use power demand, battery SOC and supercapacitor SOC as input variables, and output the power distribution ratio of battery and supercapacitor by designing fuzzy membership functions and reasoning rules.

[0008] Specifically, the strategy converts precise input variables (such as power demand and battery SOC) into fuzzy language descriptions (such as "low", "medium", "high"), and then makes decisions based on preset "if-then" rules, such as "if power demand is low and battery SOC is high, then use battery power mainly". The system converts fuzzy decisions into specific battery and supercapacitor power allocation ratios.

[0009] The above rule-based HESS control strategy has the following limitations:

[0010] 1) There is a contradiction between the rigidity of preset rules and the dynamic nature of actual applications. Since the control strategy is based on a pre-established set of rules, it is difficult to make real-time adjustments in actual applications. This rigid decision-making mechanism may result in suboptimal system performance when faced with complex and changing driving conditions. In particular, under some extreme or unexpected system conditions, the preset rules may not respond appropriately, thus affecting overall performance.

[0011] 2) It is difficult to achieve global optimization of multiple objectives. Rule-based methods usually focus on a single or limited number of optimization objectives, such as maximizing energy efficiency or minimizing power loss. However, the actual operation of HESS involves multiple interacting objectives such as energy consumption, battery life, and component performance. It is difficult to find the best balance between these objectives by relying solely on preset rules, which may result in performance improvements in some aspects at the expense of other aspects.

[0012] 3) There is a trade-off between the complexity of rule design and control accuracy. In order to improve control accuracy, it is often necessary to design a more complex and detailed set of rules. However, the increase in the number of rules not only increases the computational load of the system, but may also lead to conflicts or overlaps between rules, increasing the difficulty of strategy design and debugging. In addition, control strategies based on fixed rules are difficult to adapt to the dynamic changes of HESS performance parameters (such as battery capacity, internal resistance, etc.) with usage time and environmental conditions, which may cause the control effect to gradually deteriorate over time.

[0013] The optimization-based HESS control strategy aims to achieve the best power distribution between energy storage devices such as batteries and supercapacitors through control optimization technology. The core idea of ​​this method is to transform the control problem of the hybrid energy storage system into a typical optimization control problem. By setting appropriate objective functions (such as minimizing energy consumption, minimizing operating costs, etc.) and constraints (such as battery power limits, etc.), various optimization algorithms are used to solve the optimal control strategy. According to the optimization time scale and information utilization method, this type of strategy can be further divided into two types: global optimization method and real-time optimization method.

[0014] Based on the global optimization method, the theoretically optimal power allocation strategy is calculated for the known complete driving conditions, taking into account the energy demand and system status of the entire process. For example, Wang et al. (Wang T, Deng W, Wu J, Zhang Q. Power optimization for hybrid energy storage system of electric vehicle [C]. 2014 I EEE Conference and Expo Transport Electric Asia-Pacific (I TEC Asia-Pacific), 2014: 1-6) used the dynamic programming (DP) algorithm to minimize the total energy consumption as the objective function to optimize the power allocation of batteries and supercapacitors in the entire driving cycle. Powade et al. (Powade R, Bhateshvar Y. Design of semi-actively controlled battery-supercapacitor hybrid energy storage system[J]. Materials Today: Proceedings, 2023, 72(3): 1503-1509) transformed the HESS energy management problem into a multi-objective problem and solved it using a particle swarm optimization algorithm, while taking into account factors such as energy efficiency, battery life, and system cost.

[0015] Based on the real-time optimization method, the local optimal control decision is quickly calculated by using the instantaneous cost function instead of the global cost function for the current system state and short-term prediction information, which is suitable for real-time control in actual driving. For example, Wang Li (Wang Li. Research on Energy Management Strategy and Capacity Configuration Optimization of Hybrid Energy Storage System for Vehicles [D]. University of Science and Technology of China, 2022) uses the Model Predictive Control (MPC) method to predict the system behavior in a short period of time in the future (called the prediction time domain) based on the current system state (such as battery SOC, supercapacitor SOC, vehicle speed, etc.) at each control moment. MPC uses the system model to make multi-step predictions in this short time and obtains the optimal control sequence by solving a local optimization problem. Specifically, it takes the accumulation of instantaneous cost functions (such as instantaneous energy consumption) in the prediction time domain as the optimization target, while considering system constraints (such as power limits, etc.), so as to ensure real-time performance while also considering future impacts to a certain extent.

[0016] The above optimization-based HESS control strategy has the following limitations: 1) The practical application of the global optimization method is limited. Since the complete global driving condition information cannot be obtained in actual driving, the method based on global optimization is difficult to apply in practice. Although the driving condition for a period of time in the future can be obtained through the prediction method, there is inevitably a deviation between the predicted condition and the actual condition. This is because the actual driving environment has extremely strong randomness and uncertainty. Factors such as sudden traffic events and weather changes often cause the actual local driving condition to deviate significantly from the planned predicted condition, thereby affecting the effectiveness of the optimization results. In particular, under some extreme or unexpected system conditions, the optimization strategy based on the predicted condition may not be able to respond appropriately, thus affecting the overall performance. 2) It is difficult for the real-time optimization method to achieve economic optimization. The method based on real-time optimization usually makes short-term predictions based on the current system state and power demand, and assumes that the power demand remains constant during the deduction process. Although this simplified assumption improves the computational efficiency, it also leads to the inability to fully utilize the future condition information, making it difficult to achieve global economic optimality. In particular, when the driving condition changes rapidly, it may lead to frequent adjustments of control decisions, affecting the stability and energy efficiency of the system. Summary of the invention

[0017] The present invention proposes a HESS energy management control method based on vehicle-cloud integration, aiming to realize a control strategy combining global optimization with real-time adaptation, so as to improve the economy, energy utilization efficiency and battery life of HESS.

[0018] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0019] The HESS energy management control method based on vehicle-cloud integration includes:

[0020] The vehicle end uploads the HESS state parameters of the vehicle to the cloud; the cloud uses the real-time collected traffic information to dynamically plan the future driving conditions of the vehicle, and obtains the speed time curve and the slope time curve; sets the time value, and extracts the future driving conditions of the vehicle before the set time value in the planned future driving conditions of the vehicle as the available condition information fragment;

[0021] The cloud obtains the optimal battery SOC control sequence within the available operating condition information fragment based on the vehicle's HESS state parameters and the available operating condition information fragment;

[0022] The vehicle side manages the energy of the HESS based on the obtained optimal battery SOC control sequence.

[0023] The HESS energy management control method based on vehicle-cloud integration provided by the present invention integrates cloud-side global optimization and vehicle-side real-time control. It is different from other technical solutions in three aspects:

[0024] 1) Different from the prior art 1, the present invention uses the dynamic traffic information collected in the cloud to plan the global driving conditions, and solves the HESS optimal state sequence in the available driving condition information fragment through the DP algorithm. The vehicle side uses the HESS optimal state reference value issued to control the power output of the battery and supercapacitor, realizing the forward-looking control integrating the future driving condition information. This design can adapt to different driving conditions, and at the same time seek a balance between multiple goals such as energy consumption, battery life and component performance in the available driving condition information fragment, rather than controlling the HESS according to the optimal principle at that time, overcoming the limitations of the rule-based method that is highly rigid and difficult to achieve multi-objective global optimization.

[0025] 2) Different from the second prior art, the present invention adopts a vehicle-cloud collaborative architecture, performs global optimization in the cloud, and uses the MPC algorithm on the vehicle side for real-time adaptive adjustment. This design not only overcomes the problem that the global optimization method is difficult to obtain complete operating condition information in practical applications, but also solves the limitation that the real-time optimization method is difficult to achieve global economic optimization. By processing the power demand error caused by the uncertainty disturbance of the traffic environment and the error in predicting the driving condition in real time on the vehicle side, the present invention can effectively deal with the randomness and uncertainty in the actual driving environment, and ensure that the system can still maintain efficient and safe operation when it deviates from the planned global operating condition.

[0026] 3) Different from the prior art 1 and 2, the present invention has a dynamic weight adjustment mechanism, which can dynamically adjust the weights of the power consumption cost and the battery degradation cost according to key factors such as battery SOH, ambient temperature and system efficiency in real time. This adaptive adjustment not only takes into account the current operating status, but also takes into account the long-term performance evolution of the HESS, thereby achieving comprehensive optimization control of the entire life cycle of the system. By dynamically balancing short-term operating efficiency and long-term life extension, it effectively solves the problem that traditional methods are difficult to balance immediate benefits and long-term benefits in multi-objective optimization.

[0027] The control system of the present invention plans the global driving conditions based on dynamic traffic information in the cloud, and calculates the optimal HESS state control sequence within the available condition information fragment through a dynamic programming algorithm; the vehicle side uses the MPC algorithm to track the HESS state reference value sent by the cloud in real time, and makes local adjustments to cope with the uncertainty of the actual driving environment. This architecture achieves an effective balance between long-term optimization goals and short-term response requirements, and improves the adaptability and robustness of HESS in complex and changing environments.

[0028] The updated global driving condition front section is used as the available condition information segment, and the DP algorithm calculation is periodically performed in the cloud to obtain the optimal state control sequence of the HESS within the available condition information segment. The available condition information segment adopts a rolling update mechanism to ensure that the latest traffic information is always used during the optimization process. This dynamic optimization strategy can respond to changes in traffic conditions in a timely manner, while taking into account battery life and power consumption costs, to achieve forward-looking minimization of total costs in the future time domain.

[0029] The HESS energy management control method based on vehicle-cloud integration designed in the present invention has the following significant beneficial effects:

[0030] 1) Global optimization and forward-looking control: The global driving conditions are planned through dynamic traffic information collected from the cloud, and the DP algorithm is solved using the available condition information fragments to obtain the optimal state sequence of the HESS, so that the control strategy focuses on minimizing the total HESS cost of the entire available condition information fragment, rather than only considering the current cost. This forward-looking control ensures that the supercapacitor module can provide sufficient energy supply when the vehicle's instantaneous power demand peaks, effectively preventing the battery from high-power charging and discharging, thereby significantly improving the expected service life of the battery.

[0031] 2) Real-time adaptability and robustness: In view of the randomness and uncertainty in the actual driving environment (such as sudden traffic incidents, weather changes, etc.), the present invention uses the MPC algorithm on the vehicle side to track the battery SOC reference value sent from the cloud. This design enables the HESS to handle the power demand prediction error caused by uncertainty disturbances in real time, while taking into account the long-term and short-term benefit trade-offs of extending battery life. By realizing the organic combination of global optimization and local adjustment, the present invention can respond to environmental changes in real time and adaptively adjust the power output of the energy source, greatly improving the economy and energy efficiency of the vehicle in complex and changing environments.

[0032] 3) Dynamic weight adjustment and life cycle optimization: The adaptive weight adjustment method designed by the present invention dynamically adjusts the weights of the power consumption cost and the battery degradation cost according to the battery SOH, ambient temperature and system efficiency. This mechanism not only considers the current health status of the battery, but also takes into account the influence of the external environment and system performance, thereby achieving the optimal cost-effectiveness balance throughout the life cycle of the HESS. This comprehensive consideration ensures that the system can maintain efficient operation in different stages and environments, maximizing the economic benefits and service life of the HESS.

[0033] In summary, the energy management control method proposed in the present invention effectively solves the energy management problem of HESS in a complex and changeable environment by combining global optimization with local adjustment, forward-looking control with real-time adaptation, and dynamic weight adjustment. This multi-level and multi-dimensional control strategy not only improves the energy efficiency and economy of the vehicle, but also extends the service life of key components. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of a typical semi-active HESS configuration;

[0035] Figure 2 HESS energy management control strategy framework based on vehicle-cloud integration;

[0036] Figure 3 Schematic diagram of the rolling execution of the DP algorithm. DETAILED DESCRIPTION

[0037] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings:

[0038] HESS system configuration is as follows Figure 1 As shown. The HESS system includes a battery, a supercapacitor, and a bidirectional DC / DC inverter. The supercapacitor outputs or replenishes energy through a bidirectional DC / DC inverter, allowing the voltage of the supercapacitor to vary within a wide range. The battery is directly connected to the DC bus, so that the voltage fluctuation of the DC bus is small. HESS has the following three working modes: 1) supercapacitor alone supplies energy, 2) battery alone supplies energy, and 3) supercapacitor and battery cooperate to supply energy. The present invention designs an energy management control strategy for a typical semi-active HESS. When the vehicle is running, the HESS adjusts the energy output ratio of the battery and the supercapacitor in real time according to the driving power demand to achieve dynamic switching of different working modes. The energy output by the two is combined and superimposed, and then transmitted to the DC / AC inverter through the DC bus. The DC / AC inverter converts DC power into three-phase AC power and applies it to the drive motor. After receiving the electric energy, the motor generates torque to drive the wheels to rotate, thereby achieving continuous driving of the vehicle.

[0039] Design of HESS energy management control strategy

[0040] Establish the longitudinal dynamics model of the bus:

[0041]

[0042] Among them, i is the main reduction ratio; η mec is the mechanical transmission efficiency; T m,req is the required motor torque; R wis the wheel radius; m is the vehicle mass; g is the gravity constant; f is the rolling resistance coefficient; θ is the road slope; C D is the drag coefficient; A is the frontal area; ρ is the air density; v is the vehicle speed; δ is the rotational mass conversion coefficient; and a is the vehicle acceleration.

[0043] Build a battery dynamic model:

[0044]

[0045] in, is the battery SOC change; I bat is the battery current; Q bat is the nominal capacity of the battery; V bat is the open circuit voltage of the battery; R bat is the internal resistance of the battery; P bat Output power to the battery.

[0046] The capacity degradation Q of lithium iron phosphate battery is calculated using the semi-empirical model of capacity attenuation. loss :

[0047]

[0048] Wherein, B1 is the pre-exponential factor under 1C discharge rate condition, and its value is 27672; R represents the ideal gas constant, and its value is 8.314 J / (mol·K); T is the Kelvin temperature of the battery working environment; n is the battery discharge rate, n=I bat / Q bat ; A h1 With A hn They are respectively the accumulated ampere-hours under the condition of 1C / nC discharge rate. The absolute value of the current is taken during calculation to prevent missing the charging status during driving.

[0049] When the voltage of the supercapacitor drops to half of the maximum voltage, it no longer has the ability to discharge, so the SOC of the supercapacitor is:

[0050]

[0051] Among them, V SC is the current voltage of the supercapacitor; V max is the maximum voltage of the supercapacitor; V min is the minimum voltage of the supercapacitor;

[0052] Establish supercapacitor dynamic model:

[0053]

[0054] in, is the SOC change of the supercapacitor; I SCis the supercapacitor current; Q SC is the nominal capacity of the supercapacitor; V SC is the open circuit voltage of the supercapacitor; R SC is the internal resistance of the supercapacitor; P SC is the output power of the supercapacitor.

[0055] The actual operating conditions of buses often deviate from the preset global driving conditions due to the complex and changeable traffic environment. This deviation is caused by the random uncertainty and time-varying characteristics of the traffic environment, which makes it impossible for the vehicle to strictly follow the pre-planned driving conditions. In fact, the vehicle driving condition can be regarded as an organic combination of driving condition fragments in different time domains from the starting point to the end point. In view of this, the present invention proposes a vehicle-cloud integrated HESS energy management control strategy, which aims to improve the economic efficiency of the use of semi-active HESS, which takes into account the random change characteristics of the traffic environment and the fact that only part of the future driving condition information can be used. The control framework is as follows: Figure 2 shown.

[0056] Dynamic traffic information is collected through the intelligent transportation system. The cloud periodically plans and updates the remaining global driving conditions of the vehicle from the current position to the destination based on the collected dynamic traffic information, and obtains the speed time curve and the slope time curve. For example, the remaining global driving conditions from the current position of the vehicle to the destination are updated once every 300 seconds (not limited to 300 seconds, it can also be any value between 100 seconds and 500 seconds). This means that from the start of driving, the cloud refreshes the remaining driving conditions of the vehicle on the preset route every 300 seconds. After each update, the driving conditions of the first 300 seconds will be used as the condition information fragment available to the vehicle in the next 300 seconds. After the vehicle has driven for 300 seconds based on the available condition information fragment, the cloud updates the driving conditions using the newly collected traffic information and generates new condition information fragments for the continued driving of the vehicle.

[0057] Planning and updating the remaining global driving conditions of a vehicle from the current position to the destination is a prior art in the art, see CN108177648A. The present invention controls the bus hybrid energy storage system based on the acquired dynamically updated global driving conditions.

[0058] At the same time, the cloud uses the currently available operating condition information fragment and HESS state parameters as input and uses the bus longitudinal vehicle dynamics model to execute the DP algorithm. DP is executed every 60 seconds to solve the optimal battery SOC control sequence within the available operating condition information fragment and send the result to the vehicle as a reference value. This process continues to roll until the vehicle reaches the destination. Figure 3 Then, the vehicle uses the MPC algorithm to control the battery SOC, and tracks the reference value sent from the cloud in real time to achieve the optimal power distribution between the battery and the supercapacitor.

[0059] The optimization goal of the DP algorithm is to minimize the power consumed by the HESS internal resistance and the battery capacity attenuation cost. The corresponding objective function is:

[0060]

[0061] Among them, λ1′ and λ2′ are the modified HESS internal resistance power consumption cost weight and battery capacity attenuation cost weight respectively; k represents the time at moment k; N pre T is the end time of the available working condition information segment; S is the sampling step length; n bat is the number of battery cells; n SC is the number of supercapacitor cells; C ele is the electricity price; C bat A is the battery capacity attenuation cost; h1,20％ It is the accumulated ampere-hour when the battery capacity decays by 20% under 1C discharge condition.

[0062] Subject to system dynamics:

[0063]

[0064] and the following constraints:

[0065]

[0066] Among them, P dem,ref is the predicted vehicle driving demand power, which is obtained based on the updated global driving conditions of the vehicle from the current position to the destination; P SCDC,ref is the predicted DC / DC output power connected to the supercapacitor; P bat,ref is the predicted battery output power; η DCDC is the DC / DC efficiency; I SC,max with I SC,min are the maximum / minimum current of the supercapacitor respectively; I bat,max with I bat,min They are the maximum / minimum current of the battery; SOC SC,max With SOC SC,min They are the maximum / minimum SOC of the supercapacitor; SOC bat,max With SOC bat,min Respectively, the maximum / minimum SOC of the battery; n m is the driving motor speed; η m is the driving motor efficiency; v ref is the speed information in the available operating condition information segment; θ ref It is the slope information in the available working condition information segment.

[0067] Due to the random uncertainty of the traffic environment and the existence of prediction errors, the actual power demand of the vehicle may deviate from the predicted value on the cloud. In order to accurately respond to actual demand, the present invention determines the accurate power demand by analyzing the accelerator pedal input at the vehicle level, so that the HESS can supply energy more accurately. The MPC algorithm is used on the vehicle side to design the HESS energy management strategy, which is used to track the optimal battery SOC reference value sent from the cloud, and at the same time use the characteristics of supercapacitors to achieve "peak shaving and valley filling" to compensate for the remaining power demand fluctuations. The optimization goal of the MPC algorithm is to minimize the following error between the actual battery SOC and the reference value sent from the cloud, while minimizing the internal resistance power consumption of the HESS. In addition, MPC also takes into account the working characteristics of the battery and supercapacitor, the current SOC state, and the system constraints to ensure the stability of the energy management strategy. The objective function of the HESS energy management power allocation strategy based on MPC is:

[0068]

[0069] Among them, α represents the battery SOC tracking error weight coefficient; β and γ represent the internal resistance power consumption weight coefficients of the battery and supercapacitor respectively; N P represents the prediction time domain; k+i|k represents the prediction of the relevant state at the k+i moment at the kth moment; SOC bat,cloud Represents the optimal battery SOC reference value; R bat (SOC bat (k)) and R SC (SOC SC (k)) represent the internal resistance of the battery and the supercapacitor at the kth moment respectively.

[0070] Subject to system dynamics:

[0071]

[0072] and constraints:

[0073]

[0074] Among them, P dem Indicates the actual vehicle required power; P SCDC Indicates the output power of the DC / DC terminal connected to the supercapacitor.

[0075] Adaptive adjustment mechanism

[0076] The present invention adjusts the weight coefficients of the HESS power consumption cost and the battery attenuation cost in the cloud DP algorithm objective function (6) in real time based on the battery health state (State of Health, SOH) to achieve an adaptive balance between short-term operating costs and long-term battery life. The battery SOH is 1 in a brand new state and 0 when it reaches the end of its life, so SOH∈[0,1]. As the battery SOH decreases, the battery capacity attenuation cost weight λ2 can be gradually reduced, while the HESS internal resistance power consumption cost weight λ1 can be increased. This shows that when the battery SOH is low, the control strategy is more inclined to optimize the current operating efficiency of the HESS rather than focusing too much on extending the battery life. Based on the above adjustment strategy, the present invention adopts the following form of weight parameter adaptive adjustment mechanism:

[0077]

[0078] Among them, λ1 and λ2 are the weights of HESS internal resistance power consumption cost and battery capacity attenuation cost respectively; k1 and k2 are the weight change slopes of power consumption cost and battery attenuation cost respectively; b1 and b2 are the basic weight values ​​of power consumption cost and battery attenuation cost respectively, ensuring that reasonable weight distribution is maintained at extreme SOH values. To ensure that the sum of weight coefficients is always 1, the calculated weights are normalized:

[0079]

[0080] Among them, λ 1,norm and λ 2,norm are the normalized HESS internal resistance power consumption cost weight and battery attenuation cost weight, respectively. In addition, the present invention also considers the influence of ambient temperature (T) and HESS operating efficiency (η) on the weight, and introduces a correction factor:

[0081]

[0082] Among them, f T With f η are the correction factors for ambient temperature and HESS operating efficiency, respectively; λ3 and λ4 are the influence coefficients of temperature and efficiency, respectively; T ref With η ref are the reference temperature and reference efficiency respectively.

[0083] In summary, the modified HESS internal resistance power consumption cost weight and battery capacity attenuation cost weight adaptive adjustment calculation formula is:

[0084]

[0085] Through this adaptive mechanism, the present invention can dynamically adjust the weights of power consumption cost and battery degradation cost according to the real-time changes of battery SOH, ambient temperature and system efficiency, so as to achieve the optimal cost-benefit balance throughout the life cycle of HESS. This mechanism not only takes into account the current health status of the battery, but also takes into account the influence of the external environment and system performance.

[0086] The present invention designs an energy management control strategy for a single-motor driven pure electric bus equipped with a semi-active HESS. It should be pointed out that the energy management control strategy for the hybrid energy storage system proposed in the present invention has a wide range of applicability, not only limited to a single-motor driven system, but also applicable to multi-motor driven buses or other types of electric vehicles.

[0087] In the energy management strategy framework of the vehicle-cloud integrated hybrid energy storage system proposed in the present invention, dynamic traffic information is used in the cloud to plan and update the remaining global driving conditions, and the first 300 seconds of the updated global driving conditions are used as the available condition information fragment. It should be pointed out that the 300-second time window here is determined based on the technical ability of the current intelligent transportation system to collect dynamic information and plan global conditions. With the advancement of cloud computing technology and traffic information collection and upload technology, or due to differences in equipment in different regions, this time window may change. However, the core idea of ​​the present invention remains unchanged, that is, to use the latest traffic information to predict global conditions to achieve energy management optimization. Similarly, the setting of executing the DP algorithm every 60 seconds in the cloud is the result of a trade-off between real-time optimization and computing efficiency. This time interval may also change with the advancement of computing technology and differences in hardware equipment, but its core idea remains the same, that is, while ensuring the optimization effect, it takes into account real-time and computing resource limitations.

[0088] In the vehicle-cloud integrated HESS energy management strategy architecture proposed in the present invention, the MPC algorithm is used on the vehicle side to track the battery SOC reference value sent from the cloud, so as to realize the utilization of future operating condition information and consideration of battery life. It should be pointed out that the strategy architecture has good flexibility and scalability. For example, replacing the tracking object from the battery SOC reference value to the supercapacitor SOC reference value can also realize the effective utilization of future operating condition information and the optimization of system performance. Regardless of tracking the battery SOC or the supercapacitor SOC, the core idea of ​​the present invention is to utilize the vehicle-cloud collaborative architecture to effectively manage the HESS through predictive control methods, optimize energy distribution, and consider long-term system performance.

[0089] The adaptive adjustment method proposed in the present invention is to realize the adaptive adjustment of weight parameters in the form of a linear function combined with ambient temperature and system efficiency correction factor. It is worth noting that the adaptive adjustment mechanism of the present invention can be replaced by other methods for adjusting the weights of battery degradation cost and electric energy consumption cost in real time based on SOH changes, such as machine learning algorithms, fuzzy logic control, neural networks, etc., but the core idea remains the same, that is, dynamically adjusting the weights according to the battery SOH and other relevant factors to achieve the optimal cost-benefit balance throughout the life cycle of the HESS.

Claims

1. The HESS energy management control method based on vehicle-cloud integration is characterized by: include: The vehicle end uploads the vehicle's HESS status parameters to the cloud; The cloud uses the real-time collected traffic information to dynamically plan the vehicle's future driving conditions and obtain the speed time curve and slope time curve; A time value is set, and the future driving conditions of the vehicle within a time period before the set time value in the planned future driving conditions of the vehicle are extracted as available condition information fragments; The cloud obtains the optimal battery SOC control sequence within the available operating condition information fragment based on the vehicle's HESS state parameters and the available operating condition information fragment; The vehicle side manages the energy of HESS based on the obtained optimal battery SOC control sequence; The cloud uses the DP algorithm to solve the optimal battery SOC control sequence within the available operating condition information fragment based on the vehicle's HESS state parameters and available operating condition information fragments. The optimization goal of the DP algorithm is to minimize the power consumed by the HESS internal resistance within the available operating condition information fragment and the battery capacity attenuation cost. The corresponding objective function is: Among them, λ1′ and λ2′ are the modified HESS internal resistance power consumption cost weight and battery capacity attenuation cost weight respectively; k represents the time at the kth moment; N pre T is the end time of the available working condition information segment; S is the sampling step length; n bat is the number of battery cells; n SC is the number of supercapacitor cells; R bat (k) and R SC (k) are the internal resistance of the battery and the internal resistance of the supercapacitor at the kth moment; I bat (k) and I SC (k) are the battery current and supercapacitor current at the kth moment; n is the battery discharge rate; C ele is the electricity price; C bat A is the battery capacity attenuation cost; h1,20％ It is the accumulated ampere-hour when the battery capacity decays by 20% under 1C discharge condition. Subject to system dynamics: Among them, SOC bat (k) With SOC bat (k+1) are the battery SOC at the kth moment / k+1th moment respectively; SOC SC (k) With SOC SC (k+1) are the supercapacitor SOC at the kth moment / k+1th moment respectively; T S is the sampling period; I bat (k) and I SC (k) are the battery current and supercapacitor current at the kth moment respectively; Q bat is the nominal capacity of the battery; Q SC is the nominal capacity of the supercapacitor; V max With V min They are the maximum / minimum voltage of the supercapacitor respectively; Subject to the following constraints: P dem,ref (k)=P SCDC,ref (k)+P bat,ref (k) P bat,ref (k)=n bat (V bat (k)I bat (k)-R bat I bat (k) 2 ) I SC,min ≤I SC (k)≤I SC,max I bat,min ≤I bat (k)≤I bat,max SOC SC,min ≤SOC SC (k)≤SOC SC,max SOC bat,min ≤SOC bat (k)≤SOC bat,max Among them, P dem,ref (k) is the predicted vehicle driving demand power at the kth moment; P SCDC,ref (k) is the output power of the DC / DC terminal connected to the supercapacitor at the kth moment; P bat,ref (k) is the battery output power at the kth moment; V bat (k) and V SC (k) are the open circuit voltage of the battery and the open circuit voltage of the supercapacitor at the kth moment; I bat (k) and I SC (k) are the battery current and supercapacitor current at the kth moment respectively; R bat With R SC are the internal resistance of the battery and the internal resistance of the supercapacitor at the kth moment; SOC bat (k) With SOC SC (k) are the battery SOC and supercapacitor SOC at the kth moment respectively; η DCDC is the DC / DC efficiency; n bat is the number of battery cells; n SC is the number of supercapacitor cells; I bat,max with I bat,min are the maximum / minimum current of the battery; I SC,max with I SC,min They are the maximum / minimum current of the supercapacitor; SOC bat,max With SOC bat,min They are the maximum / minimum SOC of the battery; SOC SC,max With SOC SC,min They are the maximum / minimum SOC of the supercapacitor respectively; The cloud uses the set time period as a cycle and combines the latest HESS status information to periodically execute the DP algorithm to solve the above objective function to obtain the time from the kth moment to the end time k+N of the available operating condition information segment. pre The SOC state change sequence of the HESS that minimizes the operating cost is used as the optimal battery SOC control sequence: SOC bat (k),SOC bat (k+1),...,SOC bat (k+N pre ); After the execution of the available operating condition information fragment is completed, the cloud will generate a new available operating condition information fragment, which will be rolled over to ensure the continuous driving of the vehicle.

2. The HESS energy management control method based on vehicle-cloud integration according to claim 1 is characterized in that: The corrected HESS internal resistance power consumption cost weight λ1′ and the corrected battery capacity attenuation cost weight λ2′ are: λ1′=λ 1,norm ·f T ·f η λ2′=λ 2,norm ·f T ·f η Among them, λ 1,norm and λ 2,norm are the normalized HESS internal resistance power consumption cost weight and battery attenuation cost weight respectively; f T With f η are the correction factors for ambient temperature and HESS operating efficiency, respectively.

3. The HESS energy management control method based on vehicle-cloud integration according to claim 2 is characterized in that: The adaptive adjustment weight calculation formula is: l 1,norm =λ1 / (λ1+λ2) l 2,norm =λ2 / (λ1+λ2) Among them, λ1 and λ2 are the cost weight of HESS internal resistance power consumption and battery capacity attenuation cost respectively; λ1=k1·(1-SOH)+b1 λ2=k2·(1-SOH)+b2 Among them, k1 and k2 are the weight change slopes of power consumption cost and battery attenuation cost respectively; SOH is the battery health status; b1 and b2 are the basic weight values ​​of power consumption cost and battery attenuation cost respectively, ensuring that reasonable weight distribution is maintained at extreme SOH values; Correction factor f for ambient temperature T Correction factor f for HESS operating efficiency η They are: f T =1+λ3·(T-T ref ) / T ref f η =1+λ4·(η-η ref ) / or ref Among them, λ3 and λ4 are the influence coefficients of temperature and efficiency respectively; T ref With η ref are the reference temperature and reference efficiency respectively.

4. The HESS energy management control method based on vehicle-cloud integration according to claim 1 is characterized in that: The vehicle side performs energy management on the HESS according to the optimal battery SOC control sequence. The MPC algorithm is used to design the HESS energy management strategy on the vehicle side. The objective function of the HESS energy management power allocation strategy based on MPC is: Among them, α represents the battery SOC tracking error weight coefficient; β and γ represent the internal resistance power consumption weight coefficients of the battery and supercapacitor respectively; N P Represents the prediction time domain; SOC bat (k+i|k) indicates the battery SOC predicted at the kth moment in time k+i; SOC bat,cloud (k+i|k) represents the optimal battery SOC reference value predicted at the kth moment and at the kth moment; I bat (k+i|k) represents the battery current predicted at the kth moment in time k+i; I SC (k+i|k) represents the supercapacitor current predicted at the kth moment in time k+i; R bat (SOC bat (k)) and R SC (SOC SC (k)) represent the internal resistance of the battery and the internal resistance of the supercapacitor at the kth moment, respectively.

5. The HESS energy management control method based on vehicle-cloud integration according to claim 4 is characterized in that: The system dynamics equations and constraints for solving the objective function of the MPC-based HESS energy management power allocation strategy are: System dynamics equation: Among them, SOC bat (k) With SOC bat (k+1) are the battery SOC at the kth moment / k+1th moment respectively; I bat (k) and I SC (k) are the battery current and supercapacitor current at the kth moment respectively; T S is the sampling period; Q bat is the nominal capacity of the battery; V bat (k) and V SC (k) are the battery open circuit voltage and supercapacitor open circuit voltage at the kth moment; R bat (k) and R SC (k) are the internal resistance of the battery and the internal resistance of the supercapacitor at the kth moment; P bat (k) and P SC (k) are the battery output power and supercapacitor output power at the kth moment respectively; Constraints: P dem (k)=P bat (k)+P SCDC (k) SOC bat,min ≤SOC bat (k)≤SOC bat,max SOC SC,min ≤SOC SC (k)≤SOC SC,max I bat,min ≤I bat (k)≤I bat,max I SC,min ≤I SC (k)≤I SC,max Among them, P dem (k) is the actual vehicle power demand at the kth moment; P bat (k) is the battery output power at the kth moment; P SCDC (k) is the output power of the DC / DC terminal connected to the supercapacitor at the kth moment; SOC bat (k) With SOC SC (k) are the battery SOC and supercapacitor SOC at the kth moment; SOC bat,max With SOC bat,min They are the maximum / minimum SOC of the battery; SOC SC,max With SOC SC,min They are the maximum / minimum SOC of the supercapacitor; I bat (k) and I SC (k) are the battery current and supercapacitor current at the kth moment respectively; I bat,max with I bat,min are the maximum / minimum current of the battery; I SC,max with I SC,min They are the maximum / minimum current of the supercapacitor respectively.

6. The HESS energy management control method based on vehicle-cloud integration according to claim 1 is characterized in that: The set time value is 100-500 seconds; the future driving conditions of the vehicle planned in the previous 100-500 seconds are extracted as available driving condition information fragments.

7. The HESS energy management control method based on vehicle-cloud integration according to claim 6 is characterized in that: The set time value is 300 seconds; the future driving conditions of the vehicle planned in the previous 300 seconds are extracted as the available driving condition information fragments.

8. A HESS energy management control device, characterized in that: It includes a cloud side and a vehicle side, and the cloud side and the vehicle side perform energy management on the HESS of the vehicle according to the HESS energy management control method based on vehicle-cloud integration according to any one of claims 1 to 7; The cloud receives the HESS status parameters of the vehicle uploaded by the vehicle end; uses the real-time collected traffic information to dynamically plan the future driving conditions of the vehicle, and generates available operating condition information segments for the continuous driving of the vehicle within a set time period; obtains the optimal battery SOC control sequence within the available operating condition information segments based on the vehicle's HESS status parameters and the available operating condition information segments; and sends the optimal battery SOC control sequence to the vehicle end; On the vehicle side, the HESS status parameters of the vehicle are uploaded to the cloud; the vehicle side performs energy management on the HESS based on the obtained optimal battery SOC control sequence.

9. A vehicle, characterized in that: Energy management of the HESS of a vehicle is performed according to the HESS energy management control method based on vehicle-cloud integration as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a program or instruction, wherein the program or instruction, when executed by a processor, implements the steps of the HESS energy management control method based on vehicle-cloud integration as described in any one of 1-7.

Images