Integrated energy management method based on rule management strategy

Abstract

The invention discloses an integrated energy management method based on a rule management strategy, and the method comprises the following steps: carrying out the processing of a logic threshold rule based on the total demand power of a vehicle, the real-time state of charge (SOC) value of a power battery, and the task priority, and obtaining a working mode and a macroscopic power distribution target of a system; based on the macroscopic power distribution target, a reconfigurable power correction module is adopted for processing, and driving allowable power is obtained; based on the historical and current motion states of the vehicle, a prediction model is adopted for prediction, and a prediction power sequence is obtained; constructing an energy management optimal control problem by adopting a model predictive control method based on a predictive power sequence by taking a working mode instruction as a framework and driving allowable power as a constraint; solving an energy management optimal control problem to obtain an optimal power distribution sequence; and outputting the first-step control quantity in the optimal power distribution sequence as the final target power of the engine-generator set and the final target charging and discharging power of the power battery.

Description

Technical Field

[0001] This invention belongs to the field of hybrid vehicle control technology, and particularly relates to an integrated energy management method based on rule management strategy. Background Technology

[0002] Energy management strategies for hybrid vehicles are crucial for ensuring vehicle performance, fuel economy, and operational reliability. Currently, mainstream energy management strategies are mainly divided into two categories: rule-based and optimization-based. Rule-based energy management strategies use preset logical thresholds (such as power demand and battery state of charge) to switch operating modes and allocate power source torque. Their advantages include low computational complexity, ease of engineering implementation, and the ability to keep the engine operating in its most efficient range. Optimization-based energy management strategies, on the other hand, aim to achieve globally optimal fuel economy by establishing a global optimization model with vehicle fuel consumption or emissions as the objective function and system state as the constraint, and then using optimal control algorithms to solve the problem.

[0003] However, all of the aforementioned existing technologies have significant limitations. Rule-based strategies rely heavily on engineer experience for control parameter calibration, making it difficult to adapt to complex and ever-changing task profiles. Furthermore, they typically use fixed parameters, failing to effectively coordinate the vastly different response speeds of the front power chain (engine-generator set) and the rear power chain (battery), resulting in poor system dynamic performance and reduced energy management effectiveness during sudden power demand changes. While optimization-based strategies (such as dynamic programming) can theoretically achieve global optimization, their enormous computational burden makes them difficult to meet the real-time requirements of onboard controllers. Moreover, they typically rely on complete, pre-known operating conditions, making direct online implementation difficult in practical applications.

[0004] Therefore, this invention proposes an integrated energy management method based on a rule-based management strategy. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes an integrated energy management method based on a rule-based management strategy, thereby resolving the issues present in the prior art.

[0006] To achieve the above objectives, this invention provides an integrated energy management method based on a rule-based management strategy, comprising: Based on the vehicle's total power demand, the real-time state of charge (SOC) value of the power battery, and task priorities, logical threshold rules are used to process the data to obtain the system's operating mode and macroscopic power allocation target. Based on the aforementioned macroscopic power allocation target, and considering the differences in dynamic response characteristics between the front and rear power chains, a reconfigurable power correction module is used for processing to obtain the allowable driving power. Based on the vehicle's history and current motion state, a predictive model is used to predict the power demand in the future time domain, resulting in a predicted power sequence. Using the operating mode command as a framework and the allowable drive power as a constraint, and based on the predicted power sequence, a model predictive control method is used to construct an optimal energy management control problem with fuel economy and battery SOC balance as optimization objectives. Solving the energy management optimal control problem yields the optimal power allocation sequence between the engine-generator set and the power battery in the predicted time domain; The first control output in the optimal power allocation sequence is used as the final target power of the engine-generator set and the final target charge / discharge power of the power battery.

[0007] Optionally, the process of obtaining the system's operating mode and macroscopic power allocation target by using logical threshold rules includes: Based on the comparison of the difference between the total power demand of the vehicle and the preset optimal power pack for fuel consumption, a preliminary power allocation tendency is obtained. Based on the preliminary power allocation tendency and combined with the range of the real-time SOC value of the power battery, a multi-level threshold judgment logic is used to process the system and obtain the system's working mode. Once the operating mode is determined, based on the energy flow principle corresponding to the operating mode and combined with the current total power demand, the macroscopic power allocation target of the engine-generator set and power battery is finally generated.

[0008] Optionally, the multi-level threshold judgment logic includes: The system enters charging mode when the total power demand of the vehicle is less than the power of the fuel-efficient power pack and the battery SOC value is lower than the first preset threshold while the battery is charging. Based on the condition that the total power demand of the vehicle is less than the power of the optimal fuel consumption power pack, and the battery is not in a charging state, and the battery SOC value is higher than the second preset threshold, the system enters the optimal discharge mode. Based on the situation where the total power demand of the vehicle is greater than the power of the optimal fuel consumption power pack, the system directly enters the joint drive mode. Based on the condition that the total power demand of the vehicle is zero, the system is determined to enter the idle stop mode.

[0009] Optionally, based on the macroscopic power allocation target and considering the differences in dynamic response characteristics between the front and rear power chains, a reconfigurable power correction module is used to process and obtain the allowable driving power, including: Based on the response delay characteristics of the front power chain and the fast response characteristics of the back power chain, a dynamic response difference model of the power chain is established. Based on the dynamic response difference model, the macroscopic power allocation target output by the upper-level rule management is subjected to feedforward compensation processing to obtain the corrected front power chain power command. Based on the task priority signal, the modified power command is adjusted again to ensure that the power demand of high-priority loads is met first, and finally the allowable power of the drive is output.

[0010] Optionally, the prediction model uses a long short-term memory network to make predictions and obtain a predicted power sequence.

[0011] Optionally, the process of constructing the optimal control problem for energy management with fuel economy and battery SOC balance as optimization objectives using model predictive control methods includes: Based on the predicted power sequence, a system state equation with battery SOC as the state variable is established. Based on engine fuel consumption characteristics data and battery SOC balance requirements, a weighted objective function including fuel consumption rate and SOC deviation is constructed. An engine operating point change rate penalty term is introduced into the objective function to smooth the transition process of the engine operating point; By using the engine speed range, generator torque range, and their rate of change limits as constraints, an optimal energy management control problem is constructed.

[0012] Optionally, a dynamic programming algorithm is used to solve the optimal control problem for energy management, wherein the process of solving the control problem includes: The prediction time domain is discretized into several stages, and the battery SOC state space is discretized into several state grid points. Based on the system state equation and weighted objective function, the optimal cost function for each state at each stage is calculated using a backward recursive algorithm; and the optimal state trajectory and control sequence are found starting from the initial state using a forward search algorithm. The first control variable in the control sequence is output to the actuator as the optimal control command at the current moment.

[0013] Optionally, the method further includes a multiple constraint guarantee mechanism, which includes: Based on the physical characteristics of the engine and generator, static operating range constraints for speed and torque are set. Based on the dynamic response capabilities of the engine and generator, constraints are set on the rate of change of speed and torque. Based on the safe operating window of the power battery, set upper and lower limits of SOC and limits of charging and discharging power.

[0014] Compared with the prior art, the present invention has the following advantages and technical effects: The technical effects of this invention are significant, mainly reflected in three aspects: First, through upper-level rule management based on logical thresholds, rapid decision-making on the system's operating mode is achieved. This allows for quick switching of energy supply modes based on real-time power demand, battery SOC, and task priority, ensuring the system's real-time responsiveness and task reliability. Second, by introducing a reconfigurable power correction module, the dynamic difference between the slow response of the front power chain (engine-generator set) and the fast response of the rear power chain (power battery) is effectively compensated, smoothing power fluctuations and improving the system's stability and voltage quality under dynamic operating conditions. Finally, the lower layer employs model predictive control (MPC) for rolling optimization. Within a finite prediction time domain, precise power allocation is performed with fuel economy and battery SOC balance as objectives. Combined with dynamic programming, a near-optimal solution for global optimization is achieved, significantly reducing equivalent fuel consumption while maintaining the battery SOC within a more optimal operating range. This overall improves the hybrid system's economy, range, and energy management efficiency. Attached Figure Description

[0015] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is an energy management technology according to an embodiment of the present invention; Figure 2 This is an example of an energy management strategy based on logical thresholds in an embodiment of the present invention. Figure 3 This is an energy management strategy based on prediction and real-time rolling optimization, as described in an embodiment of the present invention. Figure 4 The power of the engine-generator set in this embodiment of the invention; Figure 5 The following are kinematic parameter curves for embodiments of the present invention, wherein (a) is a velocity curve and (b) is an acceleration curve; Figure 6 The energy management effect of this embodiment of the invention is shown in (a) as the SOC curve and (b) as the equivalent fuel consumption curve. Figure 7 The logical threshold rules are those used in embodiments of the present invention; Figure 8 This invention relates to the design of an energy management and control strategy based on MPC. Detailed Implementation

[0016] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0017] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0018] Example 1 This invention aims to provide a power balance control method for the front and rear power chains of hybrid vehicles. For distributed independent electric drive systems, power balance control of the front and rear power chains is crucial for maximizing vehicle maneuverability. This invention employs a rule-based management strategy to conduct research on operating condition prediction and demand power analysis, power coupling relationship modeling of the hybrid high-voltage system, and integrated rule-based energy management considering task priorities, breaking through the limitations of rule-based integrated hierarchical energy management technology. The control method provided by this invention can improve the safety of energy management in hybrid independent electric drive systems, providing technical support for ensuring the full realization of the maneuverability performance of hybrid vehicles.

[0019] This embodiment provides a hierarchical energy management architecture, which includes an upper-level energy management mode decision layer based on logical threshold rules and a lower-level real-time power allocation layer based on model predictive control (MPC), realizing integrated energy management of a hybrid power system. The architecture is clearly divided into two layers: the upper layer is the mode decision layer, and the lower layer is the power allocation layer. The upper layer, based on preset logical threshold rules, quickly decides the operating mode (e.g., pure electric drive, vehicle-generated electricity, combined power supply, regenerative braking, etc.) of the system based on real-time collected vehicle total demand power (Preq), current battery state of charge (SOC) value, and task priority signals (e.g., prioritizing the load of the superstructure). It then outputs mode commands and macroscopic power allocation targets. The lower layer, based on the upper-level commands, uses the model predictive control (MPC) framework to perform real-time and accurate power allocation calculations within a finite, rolling prediction time domain, with the optimization goals of improving fuel economy and maintaining battery SOC balance. It outputs the target power of the engine-generator set and the target charge / discharge power of the battery. The two layers collaborate through clear interfaces (such as allowable power and operating mode commands) to form a rapidly responsive and globally optimized integrated energy management system.

[0020] like Figure 1As shown, this embodiment provides an integrated energy management method based on a rule-based management strategy, including the following steps: Based on the vehicle's total power demand, the real-time state of charge (SOC) value of the power battery, and task priorities, logical threshold rules are used to process the data to obtain the system's operating mode and macroscopic power allocation target; based on the macroscopic power allocation target, and considering the differences in dynamic response characteristics between the front and rear power chains, a reconfigurable power correction module is used to process the data to obtain the allowable driving power; based on the vehicle's history and current motion state, a predictive model is used to predict the power demand in the future time domain to obtain a predicted power sequence; using the operating mode command as a framework and the allowable driving power as a constraint, and based on the predicted power sequence, a model predictive control method is used to construct an optimal energy management control problem with fuel economy and battery SOC balance as optimization objectives; the optimal energy management control problem is solved to obtain the optimal power allocation sequence between the engine-generator set and the power battery in the predicted time domain; the first step control quantity in the optimal power allocation sequence is output as the final target power of the engine-generator set and the final target charge / discharge power of the power battery.

[0021] The energy management control method of the present invention adopts a hierarchical control method. The upper-level control is a deterministic rule control based on logic thresholds, which determines the working mode of the engine-generator set and the power battery. The lower-level control is an energy distribution control based on model prediction, which determines the final target power of each energy supply unit.

[0022] The upper-level energy management first modifies the power supply parameters of the preceding power chain based on the reconfigurable power correction module. Then, combining the response characteristics of the preceding and following power chains, it obtains the energy supply allocation method by considering the required power value of the task priority and the current SOC value of the power battery, and outputs the allowable drive power to lay the foundation for the specific energy supply allocation ratio of the lower layer.

[0023] The upper-level energy management adopts a deterministic rule-based management strategy based on logical thresholds. This strategy has low computational complexity, simple rules, and is easy to implement in engineering. It obtains the energy supply allocation method based on the required power value and the current SOC value of the power battery. This invention comprehensively considers the vehicle's operating conditions, the power output capabilities of the power pack and battery array, and formulates specific rule-based energy management strategies, such as... Figure 2 As shown.

[0024] Based on the current vehicle speed, the current power demand and the current battery SOC value are calculated to determine the specific energy management operating mode. The specific operating mode of the logic threshold-based energy management strategy is as follows: (1)When the power required for vehicle operation is less than the power of the optimal power pack for fuel consumption, if the battery is charging, determine whether the SOC value is less than 0.8. If it is, the power pack charges the power battery. If not, the power pack discharges at the optimal power pack power.

[0025] (2)When the power required for vehicle operation is less than the power of the optimal power pack for fuel consumption, if the battery is not charging, determine whether the SOC value is greater than 0.3. If it is, the power pack discharges at the optimal power pack power. If not, the power pack charges the power battery.

[0026] (3)When the power required for vehicle operation is greater than the power of the optimal power pack for fuel consumption, discharge according to the logic threshold rule.

[0027] (4)When the power required for vehicle operation is less than 0, the system has no driving demand, and the target power of the power pack is 0.

[0028] Further, the logic threshold rule in the upper-layer energy management dynamically switches the energy supply mode, including the charging mode, the discharging mode, and the idle mode, according to the comparison result between the vehicle demand power and the optimal power pack power for fuel consumption, in combination with the SOC value of the power battery. The upper-layer energy management first corrects the power supply parameters of the front power chain according to the reconfigurable power correction module, then combines the response characteristics of the front and rear power chains, and obtains the energy supply distribution method by considering the demand power value with task priority and the current SOC value of the power battery, and outputs the driving allowable power to the drive control module, laying a foundation for the specific energy supply distribution work in the lower layer. Considering the vehicle operation conditions, the power output capabilities of the power pack and the power battery pack comprehensively, the specific logic threshold rule formulated in the present invention is as Figure 7 shown.

[0029] The core of the logic threshold rule is to compare the demand power (Preq) with a preset "optimal power pack power for fuel consumption" (P_opt), and cross-judge the interval where the SOC of the power battery is located to trigger different energy flow modes. Its specific rules include: Mode 1 (charging mode): When Preq < P_opt and the battery is in a rechargeable state (such as SOC < 0.8), control the engine-generator set to operate at the high-efficiency zone power. In addition to supplying the demand power, the excess energy charges the power battery.

[0030] Mode 2 (optimal discharging mode): When Preq < P_opt and the battery SOC is relatively high (such as SOC > 0.8) or in the discharging state, control the engine-generator set to operate at the P_opt power, and the insufficient or excessive power is adjusted by the battery (discharging or stopping charging).

[0031] Mode 3 (Combined Drive Mode): When Preq > P_opt, the engine-generator set outputs its maximum available power (P_max), and the insufficient power (Preq - P_max) is provided by the discharge of the power battery.

[0032] Mode 4 (Idle / Stop Mode): When Preq ≈ 0 (e.g., when the vehicle is stationary and there is no need for the superstructure), the engine is turned off or idled, and the required small power is provided by the battery.

[0033] The threshold values ​​in the rules (such as 0.3 and 0.8 for SOC; P_opt) can be calibrated and adjusted according to the specific vehicle model and power component parameters.

[0034] The multi-level threshold judgment logic includes: determining whether the system enters charging mode when the total power demand of the vehicle is less than the power of the optimal fuel consumption power pack, and the battery SOC value is lower than a first preset threshold while the battery is charging; determining whether the system enters optimal discharging mode when the total power demand of the vehicle is less than the power of the optimal fuel consumption power pack, and the battery SOC value is higher than a second preset threshold while the battery is not charging; determining whether the system enters combined drive mode when the total power demand of the vehicle is greater than the power of the optimal fuel consumption power pack; and determining whether the system enters idle stop mode when the total power demand of the vehicle is zero.

[0035] The reconfigurable power correction module introduced in the upper-level management is used to correct the power supply parameters of the upstream power chain and, combined with the response characteristics of the upstream and downstream power chains and task priorities, outputs the allowable drive power to the lower-level control module. This module dynamically corrects the power supply parameters of the upstream power chain (engine-generator set). It receives the original power demand command and corrects it based on the differences in the dynamic response characteristics of the upstream and downstream power chains (the engine in the upstream power chain responds slowly, while the battery in the downstream power chain responds quickly). For example, when the demand power increases rapidly, this module can instruct the upstream power chain to respond earlier or increase the target power, and then reduce the battery output after the upstream power chain has responded, thereby smoothing power fluctuations and avoiding system voltage instability or drive performance degradation caused by the delay in the upstream power chain response. This correction process also considers task priorities, ensuring that the power demands of high-priority loads are met preferentially and quickly.

[0036] Furthermore, the process of obtaining the system's operating mode and macroscopic power allocation target by using logical threshold rules includes: obtaining a preliminary power allocation tendency based on the comparison result of the difference between the vehicle's total power demand and the preset optimal fuel consumption power pack power; based on the preliminary power allocation tendency, combined with the range state of the power battery's real-time SOC value, using multi-level threshold judgment logic to obtain the system's operating mode; after the operating mode is determined, based on the energy flow principle corresponding to the operating mode, combined with the current total power demand value, the final macroscopic power allocation target of the engine-generator set and power battery is generated.

[0037] Furthermore, the process of obtaining the allowable driving power includes: establishing a dynamic response difference model of the power chain based on the response delay characteristics of the front power chain and the fast response characteristics of the back power chain; performing feedforward compensation processing on the macroscopic power allocation target output by the upper-level rule management based on the dynamic response difference model to obtain the corrected front power chain power command; and readjusting the corrected power command based on the task priority signal to ensure that the power demand of high-priority loads is met first, and finally outputting the allowable driving power.

[0038] After obtaining the vehicle's power demand in the future predicted time domain, the lower-level energy management first corrects the demand power to consider reconfigurable needs, and then achieves the underlying power allocation by constructing an optimal control problem. The optimization problem of model predictive control aims to obtain the minimum fuel consumption of the engine and maintain the balance of the power battery pack. The specific process of lower-level energy management includes: S1 deterministic rule-based high-voltage electrical system management and control strategies are relatively simple, have low computational requirements, and are easy to implement on vehicle controllers. However, the setting of their control rule parameters depends on the experience of calibration engineers.

[0039] S2 is an energy management technology based on prediction and real-time rolling optimization, namely model prediction (MPC) energy management control technology. This energy management method needs to obtain the predicted power demand in the short term. It can use LSTM to predict the operating conditions and then solve for the relatively accurate power demand of the vehicle in the predicted time domain, thereby constructing the optimal control problem.

[0040] Because the results of operating condition predictions are short-term, typically within a few seconds, and are updated in real time, general global optimization control methods (such as dynamic programming) cannot be applied to the energy management of this invention. Therefore, solving this optimal control problem requires an instantaneous optimization control method, and the MPC control method is a typical instantaneous optimization control method. The overall roadmap of the energy management strategy based on prediction and real-time rolling optimization is as follows: Figure 3 As shown.

[0041] After obtaining the vehicle's power demand in the future predicted time domain, the lower-level energy management first corrects the demand power to consider reconfigurable needs, and then achieves the underlying power allocation by constructing an optimal control problem. In the rolling optimization of model predictive control, the driving range (fuel economy) of the hybrid vehicle and the lifespan of the power battery pack are comprehensively considered to achieve optimized power allocation between the engine-generator set and the power battery. Through the front-to-rear power chain energy flow management technology of this invention, the problem of poor energy management performance caused by the slow response of the front power chain can be solved, thereby achieving integrated energy management of the front and rear power chains.

[0042] Furthermore, the Model Predictive Control (MPC) method employed in the lower-level energy management predicts the power demand in the time domain, constructs an optimization problem with the objectives of minimizing fuel consumption and balancing battery SOC, and uses a rolling optimization mechanism to achieve real-time power allocation. The MPC-based energy management control strategy framework is as follows: Figure 8 As shown. Based on the current driving state, the driving state in the future time domain is predicted. The prediction model uses the vehicle speed prediction result and applies an optimization algorithm (dynamic programming method is used in this invention) to optimize the performance index function for this period. After optimization, a set of optimal control sequences is obtained and the first step of the control quantity is applied to the system. The above steps are repeated at the next sampling time to optimize a new round of control sequences.

[0043] The specific implementation process of the MPC method is as follows: (1) Prediction. Using a pre-trained Long Short-Term Memory (LSTM) network model, based on historical vehicle speed, acceleration and other information, predict the vehicle speed sequence in the future finite time domain (such as the next 3-5 seconds), and then calculate the total demand power sequence (P_dem) in the prediction time domain. (2) Modeling. Establish an optimal control problem model that includes the system state equation (battery SOC dynamic model) and cost function. The state variable is the battery SOC, and the control variable is the engine-generator output power. (3) Optimization. In each control cycle, with the current system state as the initial value, solve the above optimal control problem to obtain a series of optimal control sequences (engine power sequences) in the prediction time domain. (4) Rolling. Only the first control quantity in the control sequence (i.e. the optimal engine target power at the current moment) is actually applied to the controlled object. In the next control cycle, update the system state, and re-predict and optimize to achieve "rolling optimization and feedback correction".

[0044] Furthermore, the dynamic programming (DP) algorithm used in the lower-level MPC is used to solve the optimal control sequence in the prediction time domain and apply the first step control quantity to the system to realize the power optimization allocation between the engine-generator set and the power battery. Discrete dynamic programming (DP) is used as the solver in each rolling optimization window of MPC. The specific steps are as follows: (1) First, the vehicle speed in the future time period is obtained according to the neural network prediction model, and the demand power in the prediction time domain is obtained through the vehicle speed in the future time period. Determine the range of the state variable SOC and divide the state variable SOC into N stages. Divide the prediction time domain into p+1 stages such as [k, k + 1, k + 2, k + 3, … , k + p]. (2) According to the dynamic programming algorithm, solve and calculate the optimal state quantity SOC and the optimal control sequence T in each stage in the prediction time domain. (3) Apply the first control quantity of the optimal control sequence as the actual output to the controlled object. (4) Push the prediction time domain forward and repeat the above steps to perform rolling optimization until the entire working condition ends.

[0045] The optimal power allocation control problem in energy management aims to minimize the cost function, i.e., to maximize driving range and extend battery life. An optimized forward dynamic programming algorithm can be designed to inversely search for the optimal state based on the stored transition states corresponding to the minimum cost, thereby obtaining the optimal power allocation strategy and the corresponding control sequence. This algorithm's cost function comprehensively considers the instantaneous fuel consumption rate and the deviation of the battery's state of charge (SOC) from the target value, adjusting the emphasis on fuel consumption and battery life through weighting factors.

[0046] Furthermore, the process of constructing the optimal control problem for energy management with fuel economy and battery SOC balance as optimization objectives includes: establishing a system state equation with battery SOC as the state variable based on the predicted power sequence; constructing a weighted objective function containing fuel consumption rate and SOC deviation based on engine fuel consumption characteristic data and battery SOC balance requirements; introducing a penalty term for the rate of change of engine operating point into the objective function to smooth the transition process of engine operating point; and constructing the optimal control problem for energy management by using the engine speed range, generator torque range and their rate of change limits as constraints.

[0047] The performance index of the optimal control problem for lower-level energy management is: In the formula, and These are the initial time and the end time, respectively. , As a weighting factor, Let be the instantaneous fuel consumption rate at time t. For the current battery , The target battery SOC.

[0048] It can be obtained from the following formula: In the formula, This is a lookup table function for the engine fuel consumption characteristic diagram determined based on bench test data.

[0049] The SOC of the power battery pack is used as the state variable of the system, and the state equation is as follows: In the formula, This is the open-circuit voltage of the power battery pack. This refers to the battery's internal resistance. For battery power, This refers to the battery's rated capacity.

[0050] In the series hybrid power system of this invention, the engine output shaft and the generator input shaft are directly connected. The engine uses speed control, and the generator uses torque control. The engine speed and generator torque must conform to the external characteristics of the engine and generator, and the changes in engine speed ne and generator torque φg between two adjacent control intervals cannot be too rapid; otherwise, the operating point cannot be transferred, and the engine-generator set cannot reach the target output power. Therefore, the following constraints must be met: In the formula, Engine speed, , These are the engine's minimum and maximum speeds, respectively. The rate of change of engine speed. , These are the minimum and maximum allowable values ​​for engine speed variations, respectively. For generator torque, , These are the generator's minimum torque and maximum torque, respectively. The generator torque variation rate, , These are the minimum and maximum allowable values ​​for generator torque variation, respectively.

[0051] The charging and discharging capacity of the power battery pack in this system is limited. To prevent overcharging and over-discharging of the power battery, the following constraints must be met: In the formula, , These are the minimum and maximum allowable SOC values ​​for the power battery pack. , These represent the maximum charging power and maximum discharging power of the power battery pack.

[0052] The multiple constraint mechanisms introduced in this invention system include limits on the rate of change of engine speed and generator torque, as well as upper and lower limits on the SOC and charging / discharging power of the power battery, to ensure system operation safety and equipment lifespan. The constraint mechanisms are used to ensure that all components of the system operate within a safe and efficient range, mainly including: (1) Dynamic constraints of power components: limiting the rate of change of engine speed (Δne / Δt) and generator torque (ΔTg / Δt) to prevent shock or control instability caused by excessively fast switching of operating points. (2) Static constraints of power components: limiting the range of engine operating speed (ne_min ~ ne_max) and generator output torque (Tg_min ~ Tg_max) to ensure that the physical external characteristics of the engine and generator are not exceeded. (3) Battery safety constraints: strictly limiting the SOC operating window of the power battery (e.g., 0.3 ~ 0.8) to prevent overcharging and over-discharging. At the same time, limiting the instantaneous maximum charging power (P_chg_max) and maximum discharging power (P_dis_max) of the battery to protect the battery health. All these constraints are introduced as hard or soft constraints when constructing the optimization problem in MPC, ensuring that the optimal control quantity obtained is feasible and safe.

[0053] This invention presents an integrated coordination mechanism for the front and rear power chains. Through the collaboration of upper-level rule management and lower-level optimization control, it addresses the response delay issue of the front power chain, achieving rapid, stable, and efficient energy flow management. This mechanism embodies the collaborative work of upper-level rule management and lower-level optimization control, and its core innovation lies in solving the problem of dynamic response mismatch between the front and rear energy sources in a hybrid power system. Upper-level rule management, through a reconfigurable power correction module, preprocesses and compensates for the front power chain's commands in the feedforward stage to address its response delay. The lower-level MPC, through feedback optimization, finely adjusts power allocation in real time to compensate for the shortcomings of feedforward control. This integrated coordination approach of "feedforward + feedback" and "macroscopic mode + microscopic optimization" ensures that the system can achieve rapid, smooth, and efficient energy flow management, whether facing slowly changing cruise conditions or sudden load increases and decreases, thereby simultaneously guaranteeing vehicle maneuverability, mission completion capability, and economy.

[0054] As a specific implementation of this embodiment, this embodiment performs real-time energy management simulation based on the NEDC cycle, and verifies the simulation on a MATLAB / Simulink energy management vehicle dynamics model. It compares indicators such as equivalent fuel consumption and power battery SOC changes using rule-based and MPC-based methods. The engine-generator set employs power control, such as... Figure 4As shown, the simulated vehicle is driving on a paved road with relatively small road surface unevenness and a large wheel adhesion rate.

[0055] The vehicle's kinematic parameters during 0-1180s are as follows: Figure 5 As shown, for this target operating condition, rule-based energy management control and model predictive energy management control are simulated respectively, and the SOC value of the power battery and the equivalent fuel consumption value are compared.

[0056] Changes in the SOC of the power battery, such as Figure 6 As shown in the figure, the blue line represents the rule-based energy management control scenario, and the red line represents the model-predictive energy management control scenario. It can be seen from the figure that, compared to traditional rule-based energy management control, the integrated energy management method based on the rule-based management strategy of this invention can maintain the SOC value in a higher range, which is more beneficial for energy management. Since the SOC value of both methods is always above the minimum discharge SOC (0.3), the balance of the power battery pack can be maintained, and energy management prioritizes minimizing equivalent fuel consumption.

[0057] The equivalent fuel consumption curve is as follows Figure 6 As shown in the figure, the black line represents the rule-based energy management control scenario, and the red line represents the model-predictive energy management control scenario. It can be seen from the figure that compared to traditional rule-based energy management control, the integrated energy management method based on the rule-based management strategy of this invention results in lower equivalent fuel consumption. Combining the power battery SOC curve and the equivalent fuel consumption curve, it can be seen that the integrated energy management method based on the rule-based management strategy of this invention performs better.

[0058] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An integrated energy management method based on rule-based management strategies, characterized in that, Includes the following steps: Based on the vehicle's total power demand, the real-time state of charge (SOC) value of the power battery, and task priorities, logical threshold rules are used to process the data to obtain the system's operating mode and macroscopic power allocation target. Based on the aforementioned macroscopic power allocation target, and considering the differences in dynamic response characteristics between the front and rear power chains, a reconfigurable power correction module is used for processing to obtain the allowable driving power. Based on the vehicle's history and current motion state, a predictive model is used to predict the power demand in the future time domain, resulting in a predicted power sequence. Using the operating mode command as a framework and the allowable drive power as a constraint, and based on the predicted power sequence, a model predictive control method is used to construct an optimal energy management control problem with fuel economy and battery SOC balance as optimization objectives. Solving the energy management optimal control problem yields the optimal power allocation sequence between the engine-generator set and the power battery in the predicted time domain; The first control output in the optimal power allocation sequence is used as the final target power of the engine-generator set and the final target charge / discharge power of the power battery.

2. The integrated energy management method based on rule-based management strategy according to claim 1, characterized in that, The process of obtaining the system's operating mode and macroscopic power allocation target by using logical threshold rules includes: Based on the comparison of the difference between the total power demand of the vehicle and the preset optimal power pack for fuel consumption, a preliminary power allocation tendency is obtained. Based on the preliminary power allocation tendency and combined with the range of the real-time SOC value of the power battery, a multi-level threshold judgment logic is used to process the system and obtain the system's working mode. Once the operating mode is determined, based on the energy flow principle corresponding to the operating mode and combined with the current total power demand, the macroscopic power allocation target of the engine-generator set and power battery is finally generated.

3. The integrated energy management method based on rule-based management strategy according to claim 2, characterized in that, The multi-level threshold judgment logic includes: The system enters charging mode when the total power demand of the vehicle is less than the power of the fuel-efficient power pack and the battery SOC value is lower than the first preset threshold while the battery is charging. Based on the condition that the total power demand of the vehicle is less than the power of the optimal fuel consumption power pack, and the battery is not in a charging state, and the battery SOC value is higher than the second preset threshold, the system enters the optimal discharge mode. Based on the situation where the total power demand of the vehicle is greater than the power of the optimal fuel consumption power pack, the system directly enters the joint drive mode. Based on the condition that the total power demand of the vehicle is zero, the system is determined to enter the idle stop mode.

4. The integrated energy management method based on rule-based management strategy according to claim 1, characterized in that, Based on the aforementioned macroscopic power allocation target, and considering the differences in dynamic response characteristics between the front and rear power chains, a reconfigurable power correction module is used to process and obtain the allowable driving power. The process includes: Based on the response delay characteristics of the front power chain and the fast response characteristics of the back power chain, a dynamic response difference model of the power chain is established. Based on the dynamic response difference model, the macroscopic power allocation target output by the upper-level rule management is subjected to feedforward compensation processing to obtain the corrected front power chain power command. Based on the task priority signal, the modified power command is adjusted again to ensure that the power demand of high-priority loads is met first, and finally the allowable power of the drive is output.

5. The integrated energy management method based on rule-based management strategy according to claim 1, characterized in that, The prediction model uses a long short-term memory network to make predictions and obtain a predicted power sequence.

6. The integrated energy management method based on rule-based management strategy according to claim 1, characterized in that, The process of constructing the optimal control problem for energy management with fuel economy and battery SOC balance as optimization objectives using model predictive control methods includes: Based on the predicted power sequence, a system state equation with battery SOC as the state variable is established. Based on engine fuel consumption characteristics data and battery SOC balance requirements, a weighted objective function including fuel consumption rate and SOC deviation is constructed. An engine operating point change rate penalty term is introduced into the objective function to smooth the transition process of the engine operating point; By using the engine speed range, generator torque range, and their rate of change limits as constraints, an optimal energy management control problem is constructed.

7. The integrated energy management method based on rule-based management strategy according to claim 6, characterized in that, The energy management optimal control problem is solved using a dynamic programming algorithm. The process of solving the control problem includes: The prediction time domain is discretized into several stages, and the battery SOC state space is discretized into several state grid points. Based on the system state equation and weighted objective function, the optimal cost function for each state at each stage is calculated using a backward recursive algorithm; and the optimal state trajectory and control sequence are found starting from the initial state using a forward search algorithm. The first control variable in the control sequence is output to the actuator as the optimal control command at the current moment.

8. The integrated energy management method based on rule-based management strategy according to claim 1, characterized in that, The method further includes a multi-constraint guarantee mechanism, which includes: Based on the physical characteristics of the engine and generator, static operating range constraints for speed and torque are set. Based on the dynamic response capabilities of the engine and generator, constraints are set on the rate of change of speed and torque. Based on the safe operating window of the power battery, set upper and lower limits of SOC and limits of charging and discharging power.

Images