Real-time power optimal distribution method and system for vehicle-mounted hybrid energy storage system

Abstract

The invention relates to the technical field of energy management of a vehicle-mounted hybrid energy storage system, and discloses a real-time power optimal distribution method and system of the vehicle-mounted hybrid energy storage system. According to the method, the future short-term operation state of the system is described by establishing a train power demand prediction model, a super-capacitor electric behavior prediction model and an electric behavior, thermal behavior and health state prediction model of a lithium battery. Under a model predictive control framework, the optimal power distribution rate is solved in real time by applying a sequential quadratic programming algorithm, and by using the method and the system, power output of a lithium battery and a super capacitor in the hybrid energy storage system can be distributed in a coordinated manner, so that the actual operation power requirement of a train is met, meanwhile, the system energy loss can be effectively reduced, and the system reliability is improved. And the temperature rise and aging of the lithium battery are inhibited, and the energy utilization efficiency and the operation safety of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of energy management technology for hybrid energy storage systems, specifically to a real-time power optimization allocation method and system for a vehicle-mounted hybrid energy storage system. Background Technology

[0002] With the rapid development of urban rail transit and energy storage equipment, on-board energy storage urban rail trains have gradually attracted widespread attention. Compared with the traditional overhead contact line power supply mode, on-board energy storage trains have many significant advantages. First, they reduce energy transmission losses in the power supply lines, improving energy utilization efficiency. Second, train operation does not rely on urban overhead power lines, thus effectively avoiding visual pollution caused by overhead contact lines and operational interruptions due to overhead contact line failures. In addition, the trains can operate flexibly on lines without overhead contact lines, bringing more convenience to the construction and operation of urban rail transit systems.

[0003] To achieve the optimal combination of high energy density and high power density, hybrid energy storage systems composed of supercapacitors and lithium batteries have become an ideal choice for urban rail trains. However, under the complex operating conditions of trains, such as frequent acceleration and deceleration and varying temperature environments, lithium batteries are highly susceptible to thermal runaway. Furthermore, excessively high temperatures and frequent high-power charging and discharging can accelerate their aging, leading to a decline in capacity and overall health. Lithium batteries also have a relatively short cycle life; as their performance degrades over time, it directly impacts train operational safety and the long-term economic benefits of the system. In addition, while supercapacitors, as auxiliary energy sources, can quickly respond to high power demands, their state of charge (SOC) limits their overuse to avoid damaging their performance and compromising the long-term operational safety of the system.

[0004] Therefore, there is an urgent need for a real-time power optimization and allocation method for onboard hybrid energy storage systems that closely relates to the actual operating conditions of trains, can effectively suppress lithium battery aging and temperature rise, and improve system energy efficiency. Summary of the Invention

[0005] The purpose of this invention is to disclose a real-time power optimization allocation method and system for vehicle-mounted hybrid energy storage systems, so as to solve the problems existing in the prior art.

[0006] To achieve the above objectives, in a first aspect, the present invention discloses a real-time power optimization allocation method for an on-board hybrid energy storage system, comprising: S1: Based on the electromagnetic torque and train speed of the train traction system collected at the current moment, predict the reference values ​​of the train speed and electromagnetic torque of the train traction control unit at the next moment; construct a train power demand prediction model based on the train speed and electromagnetic torque reference values ​​at the next moment. S2: Construct prediction models for the electrical behavior of supercapacitors, the electrical behavior of lithium batteries, and the thermal behavior of lithium batteries in vehicle-mounted hybrid energy storage systems. S3: Calculate the equivalent cumulative ampere-hour throughput of the lithium battery under the corresponding operating conditions at the current moment; calculate the ampere-hour throughput of the lithium battery from the current moment to the next moment; calculate the capacity degradation percentage of the lithium battery up to the next moment; construct a lithium battery health status prediction model based on the capacity degradation percentage; S4: Based on the power demand prediction model, calculate the predicted values ​​of lithium battery current and supercapacitor current within the prediction window; according to the predicted values ​​of lithium battery current and supercapacitor current, as well as the prediction models of supercapacitor electrical behavior, lithium battery electrical behavior, lithium battery thermal behavior, and lithium battery health status, establish relevant evaluation functions within the prediction window, including the hybrid energy storage system power loss evaluation function, the supercapacitor state of charge change evaluation function, the lithium battery temperature rise evaluation function, and the lithium battery health status degradation evaluation function; set constraints for lithium battery current, supercapacitor current, lithium battery state of charge, and supercapacitor state of charge; set weight coefficients for each evaluation function, and establish a comprehensive optimization objective function for the hybrid energy storage system based on each constraint and weight coefficient; S5: Within the prediction window, construct a standard quadratic programming form of the comprehensive optimization objective function of the hybrid energy storage system based on the comprehensive optimization objective function; and use sequential quadratic programming iterative solution to obtain the prediction sequence of the optimal power allocation rate of the lithium battery corresponding to the current control cycle. Select the first optimal power allocation rate in the optimal power allocation rate prediction sequence for the optimization control operation in the current control cycle; and reconstruct each prediction model and the comprehensive optimization objective function based on the updated state at the next moment and optimize the solution to perform rolling updates and real-time dynamic optimization of power allocation.

[0007] Secondly, this application also provides a real-time power optimization and allocation system for an on-board hybrid energy storage system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps corresponding to the method described in the first aspect above.

[0008] The present invention has the following beneficial effects: This invention employs model predictive control (MMC) to achieve real-time power optimization allocation in an onboard hybrid energy storage system. It describes the system's short-term operating state by establishing prediction models for train power demand, supercapacitor electrical behavior, and lithium battery electrical, thermal, and health status. Within the MMC framework, a sequential quadratic programming algorithm is used to solve for the optimal power allocation rate in real time. This method and system can coordinate the power output of the lithium battery and supercapacitor in the hybrid energy storage system to meet the actual operating power requirements of the train. Simultaneously, it effectively reduces system energy loss and suppresses lithium battery temperature rise and aging, thereby improving the system's energy utilization efficiency and operational safety.

[0009] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0010] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the structure of a typical on-board hybrid energy storage urban rail train power system applicable to the preferred embodiment of the present invention; Figure 2 This is a flowchart illustrating the real-time power optimization and allocation method for an on-board hybrid energy storage system according to a preferred embodiment of the present invention. Figure 3 This is a typical operating cycle speed curve of an urban rail train in a preferred embodiment of the present invention; Figure 4 This is the power demand curve corresponding to a typical operating cycle of an urban rail train in a preferred embodiment of the present invention; Figure 5 This is a comparison chart of the output power curves of lithium battery packs under the proposed optimized power allocation strategy and the traditional filter-based power allocation strategy according to a preferred embodiment of the present invention. Figure 6 This is a comparison chart of the output power curves of the supercapacitor bank under the proposed optimized power allocation strategy and the traditional filter-based power allocation strategy according to a preferred embodiment of the present invention. Figure 7 This is a comparison of lithium battery temperature change curves under the proposed optimized power allocation strategy and the traditional filter-based power allocation strategy, according to a preferred embodiment of the present invention. Figure 8 This is a comparison chart of the lithium battery health status change curves under the proposed optimized power allocation strategy and the traditional filter-based power allocation strategy in a preferred embodiment of the present invention. Figure 9This is a comparison of the state of charge change curves of a supercapacitor under the proposed optimized power allocation strategy and the traditional filter-based power allocation strategy, according to a preferred embodiment of the present invention. Figure 10 This is a comparison of the energy loss variation curves of a hybrid energy storage system under the proposed optimized power allocation strategy and the traditional filter-based power allocation strategy, according to a preferred embodiment of the present invention. Detailed Implementation The technical solution of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms "an" or "a" and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms "connected" or "linked" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up," "down," "left," "right," etc., are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship also changes accordingly.

[0012] It should be understood that the real-time power optimization and allocation method for on-board hybrid energy storage system provided in this application is applied to the power system of on-board hybrid energy storage urban rail train. The system includes an on-board hybrid energy storage system (mainly including lithium batteries, supercapacitors, and bidirectional half-bridge DC / DC converters), an intermediate DC link, and a traction system (mainly including traction inverters and traction motors) in terms of hardware structure.

[0013] like Figure 1 As shown, the preferred on-board hybrid energy storage urban rail train power system in this embodiment mainly includes: supercapacitor packs, lithium battery packs, bidirectional half-bridge DC / DC converters, intermediate DC environment, traction inverters, and traction motors.

[0014] like Figure 2 As shown, this embodiment provides a real-time power optimization allocation method for onboard hybrid energy storage urban rail trains, including: S1: Based on the electromagnetic torque and train speed of the train traction system collected at the current moment, predict the reference values ​​of the train speed and electromagnetic torque of the train traction control unit at the next moment; construct a train power demand prediction model based on the train speed and electromagnetic torque reference values ​​at the next moment. S2: Construct prediction models for the electrical behavior of supercapacitors, the electrical behavior of lithium batteries, and the thermal behavior of lithium batteries in vehicle-mounted hybrid energy storage systems. S3: Calculate the equivalent cumulative ampere-hour throughput of the lithium battery under the corresponding operating conditions at the current moment; calculate the ampere-hour throughput of the lithium battery from the current moment to the next moment; calculate the capacity degradation percentage of the lithium battery up to the next moment; construct a lithium battery health status prediction model based on the capacity degradation percentage; S4: Based on the power demand prediction model, calculate the predicted values ​​of lithium battery current and supercapacitor current within the prediction window; according to the predicted values ​​of lithium battery current and supercapacitor current, as well as the prediction models of supercapacitor electrical behavior, lithium battery electrical behavior, lithium battery thermal behavior, and lithium battery health status, establish relevant evaluation functions within the prediction window, including the hybrid energy storage system power loss evaluation function, the supercapacitor state of charge change evaluation function, the lithium battery temperature rise evaluation function, and the lithium battery health status degradation evaluation function; set constraints for lithium battery current, supercapacitor current, lithium battery state of charge, and supercapacitor state of charge; set weight coefficients for each evaluation function, and establish a comprehensive optimization objective function for the hybrid energy storage system based on each constraint and weight coefficient; S5: Within the prediction window, construct a standard quadratic programming form of the comprehensive optimization objective function of the hybrid energy storage system based on the comprehensive optimization objective function; and use sequential quadratic programming iterative solution to obtain the prediction sequence of the optimal power allocation rate of the lithium battery corresponding to the current control cycle. Select the first optimal power allocation rate in the optimal power allocation rate prediction sequence for the optimization control operation in the current control cycle; and reconstruct each prediction model and the comprehensive optimization objective function based on the updated state at the next moment and optimize the solution to perform rolling updates and real-time dynamic optimization of power allocation.

[0015] The real-time power optimization allocation method for onboard hybrid energy storage systems provided in this invention takes the onboard hybrid energy storage urban rail train power system as the research object, and deeply analyzes the application of model predictive control methods in the field of power optimization allocation. It describes the short-term future operating state of the system by establishing a train power demand prediction model, a supercapacitor electrical behavior prediction model, and lithium battery electrical behavior, thermal behavior, and health status prediction models. Under the model predictive control framework, a sequential quadratic programming algorithm is used to solve for the optimal power allocation rate in real time, ensuring that the temperature rise and aging of the lithium battery are effectively suppressed, while reducing system energy loss and improving the system's energy utilization efficiency and operational safety.

[0016] This embodiment is based on the Simulink virtual simulation platform, which consists of a lithium battery pack module, a supercapacitor pack module, a hybrid energy storage system energy management and control module, a city rail train traction system module, and a traction control unit module. This simulation platform is a commonly used existing technology in the field, and will not be described in detail here. The parameters used in the simulation experiment are shown in Table 1. Table 1 Simulation Experiment Parameters

[0017] Optionally, S1 includes: S11: Based on train longitudinal dynamics, according to The electromagnetic torque and train speed of the train traction system are collected at all times to predict... The train speed at any given time satisfies the following relationship:

[0018] In the formula, For prediction Train speed (km / h) at any time , They are respectively The electromagnetic torque and train speed of the train traction system are collected at all times. The radius of the train wheelset. The system sampling period is For the equivalent inertial mass of the train, For the load torque of the train traction system, the following relationship is satisfied:

[0019] In the formula, This is the conversion coefficient between train resistance and traction system load torque. The fitting coefficients for the 0th, 1st, and 2nd order train speeds are respectively obtained from empirical formulas. To add resistance to the train.

[0020] S12: Obtaining the train traction characteristic curve The electromagnetic torque reference value of the train traction control unit satisfies the following relationship:

[0021] In the formula, for The electromagnetic torque reference value of the train traction control unit at all times. for The speed command value of the train traction control unit at any given time. This refers to the mapping relationship between the electromagnetic torque reference value of the train traction control unit, as described by the train traction characteristic curve, and the train speed, the speed command value of the train traction control unit, and the load torque of the train traction system.

[0022] S13: Based on the prediction Based on the train speed and the electromagnetic torque reference values ​​of the train traction control unit at any given time, a train power demand prediction model is constructed, satisfying the following relationship:

[0023] In the formula, For prediction The power requirements of the train at any time , These refer to the energy transmission efficiency of the train traction inverter and the traction motor, respectively.

[0024] Optionally, S2 includes: S21: Construct a predictive model for the electrical behavior of supercapacitors in a vehicle-mounted hybrid energy storage system, satisfying the following relationship:

[0025] In the formula, For prediction The state of charge of the supercapacitor at all times. for The state of charge of the supercapacitor at all times. This refers to the input current of the supercapacitor, defined as positive during discharge and negative during charging. , These are the rated maximum open-circuit voltage and capacitance value of the supercapacitor, respectively.

[0026] S22: Construct a predictive model for the electrical behavior of lithium batteries in a vehicle-mounted hybrid energy storage system, satisfying the following relationship:

[0027] In the formula, , They are the predicted ones The voltage and state of charge across the polarized capacitor of the lithium battery at all times. , They are respectively The voltage and state of charge across the polarized capacitor of the lithium battery at all times. For lithium batteries The input current at any given time is defined as positive during discharge and negative during charging. , and These are the lithium battery capacity, polarization resistance, and polarization capacitance, respectively. S23: Construct a prediction model for the thermal behavior of lithium batteries, satisfying the following relationship:

[0028] In the formula, , They are the predicted ones Constantly monitor the core temperature and surface temperature of the lithium battery. , They are respectively Constantly monitor the core temperature and surface temperature of the lithium battery. , These represent the surface heat capacity and core heat capacity of a lithium battery, respectively. , These represent the convection resistance and thermal conductivity resistance of a lithium battery, respectively. Indicates ambient temperature. For lithium batteries The heat generation power obtained from the relationship model of the thermal effect of lithium battery electrical behavior on its internal structure satisfies the following relationship:

[0029] In the formula, for The equivalent series resistance, calculated by the model of the influence of lithium battery thermal behavior on its equivalent series resistance, is expressed by the following formula:

[0030] In the formula Indicates the equivalent series resistance of the lithium battery and k The mapping relationship between the temperature of the lithium battery at any time. for k The temperature of the lithium battery at any given time satisfies the following relationship:

[0031] Optionally, S3 includes: S31: Calculate the lithium battery's performance The equivalent cumulative ampere-hour throughput under the corresponding operating conditions at any given moment satisfies the following relationship:

[0032] In the formula, For lithium batteries The time corresponding to the operating condition is equivalent to the cutoff time. The cumulative ampere-hour throughput that produces the same capacity decay at any given time For lithium battery cutoff Percentage of capacity decay at any given time Let be the ideal gas constant. The power coefficient, and They represent The pre-exponential factor and activation energy, which are related to the lithium battery current at any given time, are calculated to satisfy the following relationship:

[0033] S32: Calculate the lithium battery's performance. Time to The throughput per ampere-hour during a given time period satisfies the following relationship:

[0034] In the formula, For lithium batteries Time to The throughput per second during a given time period.

[0035] S33: Calculate the lithium battery up to... The percentage of capacity decay at any given time satisfies the following relationship:

[0036] In the formula, For lithium battery cutoff Percentage of capacity decay at any given moment.

[0037] S34: Construct a lithium battery health state prediction model that satisfies the following relationship:

[0038] In the formula, For lithium batteries Predicting health status at all times.

[0039] Optionally, S4 includes: S41: Based on the predicted power allocation rate of the lithium battery pack and the predicted power demand of the train within the prediction window, calculate the predicted values ​​of the lithium battery current and supercapacitor current within the prediction window, satisfying the following relationship:

[0040] In the formula, , They are respectively in Predicting the future at any moment The lithium battery current and supercapacitor current values ​​are measured in steps. i For the prediction step index, , This is the total prediction step size. The prediction window is formed; Indicates in Predicting the future at any moment The power allocation rate of the lithium battery pack in each step is used as the optimization control variable of the comprehensive optimization objective function of the hybrid energy storage system. The optimal solution is obtained by iterative updating using formula (27). ; In order to be in Predicting the future at any moment The power requirements of each step train are obtained by iterative calculation using formulas (1) to (3); , These represent the number of series-connected and parallel-connected lithium battery packs in a hybrid energy storage system. , These represent the number of supercapacitors connected in series and the number of supercapacitors in the hybrid energy storage system, respectively. , They are respectively The calculation formulas for the output voltage of the lithium battery and the output voltage of the supercapacitor at any given time are as follows:

[0041] In the formula, for The open-circuit voltage of a lithium battery at any given time is expressed by the formula:

[0042] In the formula, This indicates the mapping relationship between the open-circuit voltage of a lithium battery and its temperature and state of charge.

[0043] Specifically, take Based on lithium battery pack power allocation rate sequence vector and train power demand sequence vector , respectively obtained in Prediction window for time prediction All Step-by-step lithium battery current sequence vector Supercapacitor current sequence vector .

[0044] S42: Based on the supercapacitor electrical behavior prediction model, lithium battery electrical behavior prediction model, lithium battery thermal behavior prediction model, and lithium battery health status prediction model, establish the power loss evaluation function of the hybrid energy storage system, the supercapacitor state of charge change evaluation function, the lithium battery temperature rise evaluation function, and the lithium battery health status degradation evaluation function within the prediction window, respectively.

[0045] A power loss evaluation function for a hybrid energy storage system within the prediction window is established, satisfying the following relationship:

[0046] In the formula, This represents the power loss evaluation function for a hybrid energy storage system. The equivalent series resistance of the supercapacitor in the hybrid energy storage system. This represents the maximum power loss of the hybrid energy storage system.

[0047] A function for evaluating the change in the state of charge of a supercapacitor within the prediction window is established, satisfying the following relationship:

[0048] In the formula, This represents the evaluation function for changes in the state of charge of a supercapacitor. In order to be in Time prediction The state of charge of the supercapacitor at all times.

[0049] A lithium battery temperature rise evaluation function is established within the prediction window, satisfying the following relationship:

[0050] In the formula, This represents the lithium battery temperature rise evaluation function. In order to be in Time prediction Constant lithium battery temperature.

[0051] A lithium battery health status degradation evaluation function is established within the prediction window, satisfying the following relationship:

[0052] In the formula, This represents the evaluation function for the health status degradation of lithium batteries. In order to be in Time prediction Always monitor the health status of the lithium battery. for Monitor the health status of the lithium battery at all times.

[0053] S43: Set constraints for lithium battery current, supercapacitor current, lithium battery state of charge (SOC), and supercapacitor SOC respectively, satisfying the following relationship:

[0054] In the formula, and These are the minimum and maximum allowable input currents for lithium batteries, respectively. and These are the minimum and maximum permissible states of charge for lithium batteries, respectively. and These are the minimum and maximum allowable input currents for the supercapacitor, respectively. and These represent the minimum and maximum states of charge allowed for a supercapacitor, respectively.

[0055] S44: Set the weight coefficients for each evaluation function, and in conjunction with the constraints, establish the comprehensive optimization objective function for the hybrid energy storage system, satisfying the following relationship:

[0056] In the formula, This represents the overall optimization objective function of the hybrid energy storage system. , , and These are the weighting coefficients for the power loss evaluation function, the supercapacitor state of charge change evaluation function, the lithium battery temperature rise evaluation function, and the lithium battery health status degradation evaluation function, respectively.

[0057] Optionally, S5 includes: S51: Within the prediction window, the comprehensive optimization objective function of the hybrid energy storage system is approximated twice, and the constraints are linearized to construct the standard quadratic programming form of the optimization objective function of equation (25), which satisfies the following relationship:

[0058] In the formula, The standard quadratic programming form represents the overall optimization objective function of a hybrid energy storage system. This is a sequence vector of the power allocation rate of the lithium battery pack, i.e., a sequence vector of the optimization control variables. , This is the transpose of the vector. The Hessian matrix is ​​used to optimize the objective function of the system. The vector of coefficients for linear terms. and These represent the inequality constraint matrix and inequality constraint vector, respectively, constructed after linearization of the lithium battery current constraint, supercapacitor current constraint, lithium battery state of charge constraint, and supercapacitor state of charge constraint. They are used to limit the feasible domain of the optimization variables in the prediction time domain.

[0059] S52: In the k One control cycle (system sampling cycle) (Inside), set the initial value of the lithium battery pack power allocation rate sequence vector. The sequential quadratic programming algorithm is used for iterative processing, and at the th... Within the first control cycle, the first The next iteration update satisfies the following relationship:

[0060] In the formula, , They were respectively in the second Within the first control cycle, the first , No. The sequence vector of lithium battery pack power allocation rate in the next iteration (i.e., the sequence vector of optimization control variables). In the first Within the first control cycle, the first The incremental optimization of the lithium battery pack power allocation rate sequence vector in the next iteration. In the first Within the first control cycle, the first The Hessian (second gradient) matrix of the objective function in the next iteration. Indicates the first Within the first control cycle, the first Inequality constraint matrix of the next iteration In the first Within the first control cycle, the first The corresponding constraint Lagrange multiplier vector in the next iteration Indicates the gradient calculation symbol. , This represents the total number of iterations. Specifically, the first iteration update satisfies the following relationship:

[0061] In the formula, ; The intermediate iterations are omitted here; No. The next iteration update satisfies the following relationship:

[0062] In the formula, ,and .

[0063] By iteratively solving, we obtain the result at the 1st... k One control cycle (i.e., system sampling cycle) The optimal power allocation rate sequence vector of the lithium battery pack within the prediction window is represented as follows:

[0064] In the formula, Indicates the first k One control cycle (system sampling cycle) The sequence vector of optimal power allocation rate of the lithium battery pack within the prediction window obtained by (within) Indicates the first k Within the prediction window of the control cycle, the first... Predict the optimal power allocation rate of the lithium battery pack in the prediction step. .

[0065] S53: In the k One control cycle (system sampling cycle) (Within), select the optimal power allocation rate sequence vector of the lithium battery pack within the prediction window. The optimal power allocation rate of the lithium battery pack in the first prediction step As the first k One control cycle (system sampling cycle) The optimal power allocation rate within the prediction window is used to perform optimized control operations.

[0066] Entering the next control cycle, that is, in the... One control cycle (system sampling cycle) (inside), order ,in This indicates that the value on the right side of the symbol is assigned to the value on the left side; the optimal power allocation rate sequence vector of the lithium battery pack. Assign values ​​to respectively Returning to execute S1, S2, S3, S4, and S5, we obtain the result at the [number]th [year]. New optimal power allocation rate sequence vector within each control cycle Until the machine stops.

[0067] This rolling optimization strategy enables real-time dynamic updates of the lithium battery power allocation rate, thereby effectively improving the energy utilization efficiency of the on-board hybrid energy storage system, suppressing lithium battery temperature rise, and extending its service life throughout the entire train operation process.

[0068] Specifically, the speed curve of a typical operating cycle of an urban rail train in this embodiment is as follows: Figure 3 As shown, the corresponding train power demand curve is as follows: Figure 4 As shown, the train accelerates to a given speed of 90 km / h starting at 5 seconds, begins braking at 25 seconds, enters the next operating phase at 45 seconds, accelerates to a given speed of 120 km / h, begins braking at 85 seconds, and continues until the train operation ends. This embodiment compares the proposed power optimization allocation strategy with a traditional filter-based power allocation strategy. The output power curves of the lithium battery pack under the two strategies are shown below. Figure 5 As shown; the output power curves of the supercapacitor bank under the two strategies are as follows. Figure 6 As shown; the temperature change curves of the lithium battery under the two strategies are as follows. Figure 7 As shown; the health status change curves of lithium batteries under the two strategies are as follows. Figure 8 As shown; the charge state change curves of the supercapacitor under the two strategies are as follows. Figure 9 As shown; the energy loss variation curves of the hybrid energy storage system under the two strategies are as follows: Figure 10As shown. It should be noted that in this embodiment, the initial state of health of the lithium battery is set to 98%, the ambient temperature is set to 25 ℃, and the initial state of charge (SOC) of the supercapacitor is set to 75%.

[0069] In summary, this application first constructs a power demand prediction model based on the train traction characteristic curve to obtain the predicted train power demand; secondly, it establishes prediction models for the electrical behavior of supercapacitors, the electrical behavior of lithium batteries, and the thermal behavior of batteries to obtain the predicted key electrical and thermal state parameters of the on-board hybrid energy storage system; then, it calculates the equivalent cumulative ampere-hour throughput, capacity degradation percentage, and health status of the lithium battery to construct a lithium battery health status prediction model to obtain the predicted lithium battery health status; further, it establishes a system comprehensive optimization objective function, including power loss, supercapacitor state of charge change, lithium battery temperature rise, and health status degradation evaluation functions, based on each prediction model and constraint conditions; finally, within the prediction window, it approximates the optimization objective function quadratically and linearizes the constraints, then iteratively solves the problem using a sequential quadratic programming algorithm to obtain the optimal lithium battery pack power allocation rate prediction sequence for the current control cycle. In each control cycle, only the first optimization decision in this sequence is executed, and the system power allocation is updated in real time in the next control cycle through a rolling optimization strategy. This effectively improves the energy utilization efficiency of the on-board hybrid energy storage system, suppresses lithium battery temperature rise, and extends its service life throughout the entire train operation process. Corresponding to the above method embodiments, the real-time power optimization and allocation system for a vehicle-mounted hybrid energy storage system provided in this embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above method. This real-time power optimization and allocation system for a vehicle-mounted hybrid energy storage system can implement all embodiments of the above-described real-time power optimization and allocation method for a vehicle-mounted hybrid energy storage system and achieve the same beneficial effects, which will not be elaborated here.

[0070] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A real-time power optimization distribution method for a vehicle-mounted hybrid energy storage system, characterized in that, The method comprises the following steps: S1: predicting the train speed and the electromagnetic torque reference value of the train traction control unit at the next moment according to the train traction system electromagnetic torque and the train speed collected at the current moment; constructing a train power demand prediction model according to the train speed and the electromagnetic torque reference value at the next moment; S2: constructing a supercapacitor electrical behavior prediction model of the on-board hybrid energy storage system, a lithium battery electrical behavior prediction model of the on-board hybrid energy storage system, and a lithium battery thermal behavior prediction model of the on-board hybrid energy storage system; S3: calculating the equivalent cumulative ampere-hour throughput of the lithium battery under the working condition corresponding to the current moment; calculating the ampere-hour throughput of the lithium battery during the period from the current moment to the next moment; calculating the capacity degradation percentage of the lithium battery up to the next moment; and constructing a lithium battery state of health prediction model according to the capacity degradation percentage; S4: calculating the lithium battery current and supercapacitor current prediction values in the prediction window based on the power demand prediction model; establishing relevant evaluation functions in the prediction window according to the lithium battery current, supercapacitor current prediction values, and the supercapacitor electrical behavior prediction model, lithium battery electrical behavior prediction model, lithium battery thermal behavior prediction model, and lithium battery state of health prediction model, wherein the relevant evaluation functions comprise a hybrid energy storage system power loss evaluation function, a supercapacitor state of charge change evaluation function, a lithium battery temperature rise evaluation function, and a lithium battery state of health degradation evaluation function; and setting lithium battery current, supercapacitor current, lithium battery state of charge, and supercapacitor state of charge constraint conditions; setting weight coefficients of each evaluation function, and establishing a hybrid energy storage system comprehensive optimization objective function based on each constraint condition and each weight coefficient; S5: constructing a standard quadratic programming form of the hybrid energy storage system comprehensive optimization objective function in the prediction window based on the comprehensive optimization objective function; and solving the standard quadratic programming form by using a sequential quadratic programming iteration to obtain a lithium battery optimal power distribution rate prediction sequence corresponding to the current control period, selecting the first optimal power distribution rate in the optimal power distribution rate prediction sequence for performing an optimal control operation in the current control period; and reconstructing each prediction model and the comprehensive optimization objective function based on an updated state at the next moment and optimizing the reconstruction to perform rolling update and real-time dynamic optimization of power distribution.

2. The real-time power optimization and distribution method for a vehicle hybrid energy storage system of claim 1, wherein, The S1 comprises: S11: based on the longitudinal dynamics of the train, according to the electromagnetic torque of the train traction system and the train speed collected at the moment, the train speed at the moment is predicted the train speed at the moment, the following relationship is satisfied: In the formula, is the predicted train speed at the moment, , respectively, electromagnetic torque of the train traction system and train speed collected at the moment, is the train wheel pair radius, is the system sampling period, is the equivalent inertial mass of the train, is the train traction system load torque, and satisfies the following relationship: In the formula, is the conversion coefficient between the train resistance and the load torque of the traction system, are the fitting coefficients of the 0th, 1st and 2nd order train rotating speeds respectively, which are obtained from an empirical formula, is the additional resistance of the train; S12: Obtain the train traction characteristic curve, and obtain The time electromagnetic torque reference value of the train traction control unit satisfies the following relationship: wherein is the electromagnetic torque reference value of the train traction control unit at the moment, is the speed command value of the train traction control unit at the moment, is a mapping relationship between the electromagnetic torque reference value of the train traction control unit described by the train traction characteristic curve and the train speed, the speed command value of the train traction control unit, and the train traction system load torque. S13: according to the prediction At the moment of train speed and train traction control unit electromagnetic torque reference value, the construction of train power demand prediction model, meet the following relationship: In the formula, is predicted train power demand at the moment, , respectively the energy transmission efficiency of the train traction inverter, the traction motor.

3. The real-time power optimization and distribution method for a vehicle hybrid energy storage system of claim 1, wherein, The S2 comprises: S21: constructing a supercapacitor electrical behavior prediction model of the on-board hybrid energy storage system, which satisfies the following relationship: wherein, is the predicted is the state of charge of the supercapacitor at time is the state of charge of the supercapacitor at time is the state of charge of the supercapacitor at time is the supercapacitor input current, defined as positive when discharging and negative when charging, , are the rated maximum open circuit voltage and capacitance of the supercapacitor, respectively, is the system sampling period; S22: constructing a lithium battery electrical behavior prediction model of the on-board hybrid energy storage system, which satisfies the following relationship: wherein, , are the predicted voltage across the polarization capacitor and state of charge of the lithium battery at time , are the predicted voltage across the polarization capacitor and state of charge of the lithium battery at time is the input current of the lithium battery at time , defined as positive for discharging and negative for charging, , and are the capacity, polarization resistance and polarization capacitance of the lithium battery, respectively. S23: constructing a lithium battery thermal behavior prediction model, which satisfies the following relationship: wherein, , are the predicted lithium battery core temperature, surface temperature at time , are the lithium battery core temperature, surface temperature at time , denote the lithium battery surface heat capacity, core heat capacity, , denote the lithium battery convection resistance, thermal conduction resistance, denotes the ambient temperature, is the heat generation power of the lithium battery at time according to the relationship model of the thermal influence on the inside of the lithium battery according to the electrical behavior of the lithium battery, satisfying the following relationship: In the formula, is The equivalent series resistance calculated by the mapping relationship model of the influence of the thermal behavior of the lithium battery on the equivalent series resistance meets the following relationship: In the formula represents the equivalent series resistance of the lithium battery and k the mapping relationship between the temperature of the lithium battery at the moment and is k the temperature of the lithium battery at the moment, satisfying the following relationship: （10）。 4. The real-time power optimization and distribution method for a vehicle hybrid energy storage system of claim 1, wherein, The S3 comprises: S31: Calculate the equivalent cumulative ampere-hour throughput of the lithium battery at the time corresponding to the working condition, which satisfies the following relationship: S31: Calculate the equivalent cumulative ampere-hour throughput of the lithium battery at the time corresponding to the working condition, which satisfies the following relationship: In the formula, For lithium batteries The time corresponding to the operating condition is equivalent to the cutoff time. The cumulative ampere-hour throughput that produces the same capacity decay at any given time For lithium battery cutoff Percentage of capacity decay at any given time. Let be the ideal gas constant. The power coefficient is and They represent The pre-exponential factor and activation energy, which are related to the lithium battery current at any given time, are calculated to satisfy the following relationship: In the formula, for lithium batteries in the input current at the moment S32: Calculate the lithium battery's performance. Time to The throughput per ampere-hour during a given time period satisfies the following relationship: wherein is the lithium battery capacity, is the lithium battery capacity, is the ampere-hour throughput of the lithium battery during the time period from is the lithium battery capacity, is the system sampling period; S33: Calculate the lithium battery up to... The percentage of capacity decay at any given time satisfies the following relationship: wherein is the capacity fade percentage at time t is the capacity fade percentage at time t S34: constructing a lithium battery state of health prediction model, which satisfies the following relationship: In the formula, for lithium batteries in momentary predicted state of health.

5. The real-time power optimization and distribution method for a vehicle hybrid energy storage system of claim 1, wherein, The S4 comprises: S41: calculating lithium battery current and supercapacitor current prediction values in the prediction window based on the lithium battery group power distribution rate and train power demand prediction values in the prediction window, which satisfy the following relationship: In the formula, , are the predicted future lithium battery current, super capacitor current values of the future 1th step at the time of , is the prediction step index, i , , is the total prediction step length, is the prediction window composed of; represents the predicted future lithium battery power distribution rate of the future 1th step at the time of , is used as an optimization control variable of the hybrid energy storage system comprehensive optimization objective function, ; is the predicted future train power demand of the future 1th step at the time of , , , are the series number and parallel number of the lithium battery group of the hybrid energy storage system, , are the series number and parallel number of the super capacitor group of the hybrid energy storage system; , are the lithium battery output end voltage and super capacitor output end voltage at the time of , the calculation formulas are as follows: In the formula, is The open-circuit voltage of the lithium battery at the moment t is given by the formula In the formula, represents the mapping relationship between the open-circuit voltage of the lithium battery and the temperature and state of charge of the lithium battery. S42: according to the super capacitor electrical behavior prediction model, lithium battery electrical behavior prediction model, lithium battery thermal behavior prediction model and lithium battery health state prediction model, respectively establish the mixed energy storage system power loss evaluation function, super capacitor state of charge change evaluation function, lithium battery temperature rise evaluation function, lithium battery health state degradation evaluation function in the prediction window: The mixed energy storage system power loss evaluation function in the prediction window is established, which satisfies the following relationship: In the formula, represents the power loss evaluation function of the hybrid energy storage system, is the equivalent series resistance of the super capacitor of the hybrid energy storage system, is the maximum value of the power loss of the hybrid energy storage system, is the polarization resistance; The super capacitor state of charge change evaluation function in the prediction window is established, which satisfies the following relationship: wherein represents a supercapacitor state of charge variation evaluation function, is a prediction of the supercapacitor state of charge at the moment; The lithium battery temperature rise evaluation function in the prediction window is established, which satisfies the following relationship: In the formula, represents a lithium battery temperature rise evaluation function, is the lithium battery temperature at time predicted at time is the lithium battery temperature at time The lithium battery health state degradation evaluation function in the prediction window is established, which satisfies the following relationship: In the formula, represents a lithium battery health state degradation evaluation function, is a lithium battery health state at a time point, is a lithium battery health state at a time point, is a lithium battery health state at a time point, is a lithium battery health state at a time point, is a lithium battery health state at a time point; S43: lithium battery current constraint condition, super capacitor current constraint condition, lithium battery state of charge constraint condition and super capacitor state of charge constraint condition are set respectively, which satisfy the following relationship: wherein, and are the minimum and maximum input current allowed by the lithium battery, respectively, and are the minimum and maximum state of charge allowed by the lithium battery, respectively, and are the minimum and maximum input current allowed by the supercapacitor, respectively, and are the minimum and maximum state of charge allowed by the supercapacitor, respectively. S44: the weight coefficients of each evaluation function are set, the mixed energy storage system comprehensive optimization objective function is established combined with the constraint conditions, which satisfy the following relationship: In the formula, denotes the comprehensive optimization objective function of the hybrid energy storage system, , , and are weight coefficients of the set power loss evaluation function, the super capacitor state of charge change evaluation function, the lithium battery temperature rise evaluation function, and the lithium battery state of health degradation evaluation function, respectively.

6. The real-time power optimization and distribution method for a vehicle hybrid energy storage system of claim 1, wherein, The S5 includes: S51: in the prediction window, the mixed energy storage system comprehensive optimization objective function is twice approximated, the constraint conditions are linearly processed, the standard quadratic programming form of the optimization objective function is constructed, which satisfies the following relationship: wherein, represents a standard quadratic programming form of the hybrid energy storage system comprehensive optimization objective function, is a lithium battery power distribution rate sequence vector, , is a transpose of a vector, is a Hessian matrix of the system comprehensive optimization objective function, is a linear term coefficient vector, and respectively represent an inequality constraint matrix and an inequality constraint vector constructed after linearization of a lithium battery current constraint, a super capacitor current constraint, a lithium battery state of charge constraint, and a super capacitor state of charge constraint. S52: In the first control cycle, set the initial value of the lithium battery power distribution rate sequence vector , using the sequence quadratic programming algorithm for successive iteration, in the first control cycle, the first iteration updates the value that satisfies the following relationship: ​​​ In the formula, , They were respectively in the second Within the first control cycle, the first , No. The sequence vector of lithium battery pack power allocation rate in the next iteration. In the first Within the first control cycle, the first The incremental optimization of the lithium battery pack power allocation rate sequence vector in the next iteration. In the first Within the first control cycle, the first The Hessian matrix of the objective function in the next iteration Indicates the first Within the first control cycle, the first Inequality constraint matrix of the next iteration In the first Within the first control cycle, the first The corresponding constraint Lagrange multiplier vector in the next iteration Indicates the gradient calculation symbol. , This represents the total number of iterations. By iterative solution, the optimal power distribution rate sequence vector of the lithium battery pack in the first control cycle prediction window is obtained, satisfying the following relationship: wherein represents the optimal power distribution rate sequence vector of the lithium battery pack within the prediction window obtained at the th control cycle, represents the optimal power distribution rate value of the lithium battery pack at the th step of the prediction at the th control cycle, ; S53: In the In each control cycle, the optimal power allocation rate sequence vector of the lithium battery pack within the prediction window is selected. The optimal power allocation rate of the lithium battery pack in the first prediction step As the first The optimal power allocation rate within a control cycle prediction window is used to perform optimized control operations. The next control cycle, i.e. the first control cycle, is entered, and , wherein denotes that the quantity to the right of the symbol is assigned to the quantity to the left; the lithium battery pack optimal power distribution rate sequence vector is assigned to , respectively, and execution is returned to S1, S2, S3, S4, S5, resulting in a new optimal power distribution rate sequence vector for the first control cycle; until shutdown.

7. A real-time power optimization distribution system for a vehicle hybrid energy storage system, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, The processor executes the computer program to realize the steps of the method in any one of claims 1-6.

Images