Energy data processing system of hydrogen electricity HIL model based on dynamic domain controller

Abstract

The invention relates to an energy data processing system of a hydrogen electricity HIL model based on a dynamic domain controller, and the system comprises a hydrogen pile energy supply module, a charging energy supply module, a motor consumption module, a comprehensive energy calculation module, a dynamic vehicle module, a working condition selection module, a speed parameter processing module, and a slope parameter processing module. The acquisition module is used for acquiring data information of supply energy of a hydrogen reactor in real time; the charging energy supply module is used for acquiring data information of supply energy of a power battery in real time; and the motor consumption module is used for acquiring data information of consumed energy of the motor of the whole vehicle in real time. According to the invention, the dynamic nature of the whole energy calculation is increased by the hydrogen electricity HIL model, and the real vehicle data is imported into the working condition selection module, so that the data authenticity is improved, the parameter realization is classified, and the operation diversification of the HIL model is improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to an energy data processing system of a hydrogen-electric HIL model based on a power domain controller. Background Art

[0002] Commercial vehicles have higher requirements for endurance and more complex functions than passenger vehicles. In the process of developing power domain controllers, the hardware-in-the-loop test requirements of the controllers pose a considerable challenge. The application of hardware-in-the-loop (HIL) helps to restore the actual vehicle environment and various operating conditions, and can effectively solve the problem of test efficiency. It is the key to realizing controller model development. However, due to the diversification of commercial vehicle power technology development, the current sources of commercial vehicle power domains are mainly concentrated in the fields of pure electric / oil-electric, and most solutions are mainly extended-range or plug-in. There is basically no progress in the HIL test model in the hydrogen-electric field.

[0003] Although there are many HIL model schemes for the three-electric power system of pure electric commercial vehicles, the controlled object model of hydrogen electric commercial vehicles is quite different from that of pure electric / fuel-powered commercial vehicles. The design of the pure electric test model is not completely suitable for hydrogen fuel cell vehicles.

[0004] However, with the development of diversified power of commercial vehicles, the application requirements of relatively complex working conditions and roads of power systems and convenient model development, the hydrogen fuel HIL test model will be the key point in the development of commercial vehicle hydrogen electric power system models. The HIL model will use a large number of working condition data points. How to efficiently run the controlled object model and solve the difficulties of multi-condition switching operation will be a new challenge we face. Summary of the invention

[0005] In view of the above problems, the present invention provides an energy data processing system of a hydrogen-electric HIL model based on a power domain controller. Not only does the hydrogen-electric HIL model increase the dynamics of the entire energy calculation, but the operating condition selection module also imports real vehicle data to improve data authenticity, categorizes parameters, and enhances the diversity of HIL model operation.

[0006] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows: an energy data processing system of a hydrogen-electric HIL model based on a power domain controller, the system comprising a hydrogen stack energy supply module, a charging energy supply module, a motor consumption module, a comprehensive energy calculation module, a dynamic vehicle module, a working condition selection module, a speed parameter processing module and a slope parameter processing module,

[0007] The hydrogen stack energy supply module is used to obtain data information of the energy supplied by the hydrogen stack in real time;

[0008] The charging energy supply module is used to obtain data information of energy supplied by the power battery in real time; the motor consumption module is used to obtain data information of energy consumed by the motor of the whole vehicle in real time; the comprehensive energy calculation module is connected to the hydrogen stack energy supply module, the power supply module and the motor consumption module, and is used to predict the comprehensive energy loss of the whole vehicle based on the data information of energy supplied by the hydrogen stack, the data information of energy supplied by the power battery and the data information of energy consumed by the motor of the whole vehicle, using the BP neural network regression prediction algorithm based on ridge regression optimization to obtain the predicted data information of the comprehensive energy loss of the whole vehicle;

[0009] The dynamic vehicle module is connected to the comprehensive energy calculation module, and is used to build a whole vehicle energy management model based on the vehicle dynamics model according to the predicted data information of the comprehensive energy loss of the whole vehicle and the usage data required by each actuator, so as to manage the energy of the whole vehicle and output the data information of the energy management of the whole vehicle.

[0010] Furthermore, the speed parameter processing module is used to process the data information of speed type according to the change of vehicle speed, and the slope parameter processing module is used to process the data information of slope type according to the change of slope.

[0011] Furthermore, the working condition selection module is connected with the speed parameter processing module, the slope parameter processing module and the dynamic vehicle module, and is used to construct a working condition category selection function Q according to the data information of the energy management of the whole vehicle, the data information of the speed type, the data information of the slope type and the data information of the real vehicle of different cycle working conditions imported,

[0012]

[0013] Among them, x1 is the data information of the energy management of the whole vehicle, x2 is the data information of the speed type, x3 is the data information of the slope type, x4 is the data information of the actual vehicle under different imported cycle conditions, α1, α2 and α3 are the decision factors of the working condition category, which characterize the category selection of the working condition and obtain the data information of the working condition selection category.

[0014] Furthermore, the decision factors α1, α2 and α3 of the working condition category are:

[0015]

[0016] Among them, x1 is the data information of the energy management of the whole vehicle, x2 is the data information of the speed type, x3 is the data information of the slope type, and x4 is the data information of the actual vehicle under different imported cycle conditions.

[0017] Furthermore, the working condition selection module is connected to a host computer working condition display module, and the host computer working condition display module is used to display data information of the selected category of the working condition in real time.

[0018] Furthermore, the method of using the BP neural network regression prediction algorithm based on ridge regression optimization to predict the comprehensive energy loss of the whole vehicle includes:

[0019] L1. Inputting the energy supply data information of the hydrogen stack, the energy supply data information of the power battery and the energy consumption data information of the vehicle motor into the BP neural network, initializing the weight and bias of the BP neural network, and obtaining the weight and bias data information of the initialized BP neural network;

[0020] L2. Based on the weight and bias data information of the initialized BP neural network, a ridge regression optimization function W is established.

[0021]

[0022] Among them, y1 is the data information of the weight of the initialized BP neural network, y2 is the data information of the bias of the initialized BP neural network, β1, β2 and β3 are penalty coefficients, and the weight and bias of the BP neural network are optimized to obtain the optimized BP neural network;

[0023] L3. Based on the optimized BP neural network, the data information of the energy supply of the hydrogen stack, the data information of the energy supply of the power battery and the data information of the energy consumption of the vehicle motor are input to determine the regression prediction function R of the BP neural network,

[0024]

[0025] Among them, r1 is the data information of the energy supplied by the hydrogen stack, r2 is the data information of the energy supplied by the power battery, r3 is the data information of the energy consumed by the vehicle motor, δ1, δ2 and δ3 are regression prediction factors, which predict the comprehensive energy loss of the vehicle and obtain the predicted data information of the comprehensive energy loss of the vehicle.

[0026] Furthermore, the constraints of the regression prediction factors δ1, δ2 and δ3 are:

[0027]

[0028] Furthermore, the penalty coefficients β1, β2 and β3 are,

[0029]

[0030] Among them, y1 is the data information of the weight of the initialized BP neural network, and y2 is the data information of the bias of the initialized BP neural network.

[0031] Furthermore, the construction of a vehicle energy management model based on the vehicle dynamics model to manage the energy of the vehicle includes:

[0032] U1. Based on the predicted comprehensive energy loss data of the vehicle and the usage data required by each actuator, establish a relationship function P between energy loss and actuator usage,

[0033] Among them, z1 is the data information of the predicted comprehensive energy loss of the whole vehicle, z2 is the usage data information required to receive each actuator, γ1, γ2 and γ3 are relationship constant parameters, which characterize the relationship between the energy loss of the whole vehicle and the energy required by the actuator, and obtain the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator;

[0034] U2. Input the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator into the whole vehicle energy management model based on the vehicle dynamics model for training and learning, and determine the whole vehicle energy management function S,

[0035]

[0036] Among them, h is the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator, η1, η2 and η3 are the allocation factors of the whole vehicle energy management, and the trained whole vehicle energy management model based on the vehicle dynamics model is obtained;

[0037] U3. Based on the trained vehicle energy management model based on the vehicle dynamics model, the predicted data information of the comprehensive energy loss of the whole vehicle and the usage data required by each actuator are input, the energy of the whole vehicle is managed, and the data information of the energy management of the whole vehicle is output.

[0038] Furthermore, the constraint function f of the allocation factors η1, η2 and η3 of the vehicle energy management is,

[0039]

[0040] The value range of the constraint function f is (0,1).

[0041] The present invention has the following positive effects:

[0042] 1. The present invention predicts the comprehensive energy loss of the whole vehicle by adopting a BP neural network regression prediction algorithm based on ridge regression optimization, obtains the data information of the predicted comprehensive energy loss of the whole vehicle, and combines it with the construction of a whole vehicle energy management model based on the vehicle dynamics model to manage the energy of the whole vehicle. It can not only accurately control and manage the energy loss of the whole vehicle, but also dynamically predict the energy consumption of the whole vehicle, thereby improving the efficiency of vehicle operation and reducing the energy consumption of the whole vehicle.

[0043] 2. The working condition selection module of the present invention imports real vehicle data, improves data authenticity, categorizes parameters, and enhances the diversification of HIL model operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0045] Figure 2 It is a schematic diagram of the process of the BP neural network regression prediction algorithm based on ridge regression optimization of the present invention;

[0046] Figure 3 The present invention is a flow chart of constructing a vehicle energy management model based on a vehicle dynamics model. DETAILED DESCRIPTION

[0047] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0048] Example 1: Figure 1 As shown, an energy data processing system of a hydrogen-electric HIL model based on a power domain controller includes a hydrogen stack energy supply module, a charging energy supply module, a motor consumption module, a comprehensive energy calculation module, a dynamic vehicle module, a working condition selection module, a speed parameter processing module and a slope parameter processing module.

[0049] The hydrogen stack energy supply module is used to obtain data information of the energy supplied by the hydrogen stack in real time;

[0050] The charging energy supply module is used to obtain data information on the energy supplied by the power battery in real time; the motor consumption module is used to obtain data information on the energy consumed by the motor of the whole vehicle in real time; the comprehensive energy calculation module is connected to the hydrogen stack energy supply module, the power supply module and the motor consumption module, and is used to predict the comprehensive energy loss of the whole vehicle based on the data information on the energy supplied by the hydrogen stack, the data information on the energy supplied by the power battery and the data information on the energy consumed by the motor of the whole vehicle, using a BP neural network regression prediction algorithm based on ridge regression optimization to obtain the predicted data information on the comprehensive energy loss of the whole vehicle; the dynamic vehicle module is connected to the comprehensive energy calculation module, and is used to construct a whole vehicle energy management model based on the vehicle dynamics model according to the predicted data information on the comprehensive energy loss of the whole vehicle and the usage data required by each actuator, so as to manage the energy of the whole vehicle and output the data information on the energy management of the whole vehicle.

[0051] In this embodiment, the speed parameter processing module is used to process the data information of speed type according to the change of vehicle speed, and the slope parameter processing module is used to process the data information of slope type according to the change of slope.

[0052] In this embodiment, the operating condition selection module is connected to the speed parameter processing module, the slope parameter processing module and the dynamic vehicle module, and is used to construct an operating condition category selection function Q according to the data information of the energy management of the whole vehicle, the data information of the speed type, the data information of the slope type and the data information of the real vehicle of different cycle operating conditions imported,

[0053] Among them, x1 is the data information of the energy management of the whole vehicle, x2 is the data information of the speed type, x3 is the data information of the slope type, x4 is the data information of the actual vehicle under different imported cycle conditions, α1, α2 and α3 are the decision factors of the working condition category, which characterize the category selection of the working condition and obtain the data information of the working condition selection category.

[0054] In this embodiment, the decision factors α1, α2 and α3 of the operating condition category are:

[0055]

[0056] Among them, x1 is the data information of the energy management of the whole vehicle, x2 is the data information of the speed type, x3 is the data information of the slope type, and x4 is the data information of the actual vehicle under different imported cycle conditions.

[0057] In this embodiment, the working condition selection module is connected to the working condition display module of the host computer, and the working condition display module of the host computer is used to display data information of the selected category of the working condition in real time.

[0058] In this embodiment, if Figure 2 As shown, the prediction of the comprehensive energy loss of the whole vehicle using the BP neural network regression prediction algorithm based on ridge regression optimization includes:

[0059] L1. Inputting the energy supply data information of the hydrogen stack, the energy supply data information of the power battery and the energy consumption data information of the vehicle motor into the BP neural network, initializing the weight and bias of the BP neural network, and obtaining the weight and bias data information of the initialized BP neural network;

[0060] L2. Based on the weight and bias data information of the initialized BP neural network, a ridge regression optimization function W is established.

[0061]

[0062] Among them, y1 is the data information of the weight of the initialized BP neural network, y2 is the data information of the bias of the initialized BP neural network, β1, β2 and β3 are penalty coefficients, and the weight and bias of the BP neural network are optimized to obtain the optimized BP neural network;

[0063] L3. Based on the optimized BP neural network, the data information of the energy supply of the hydrogen stack, the data information of the energy supply of the power battery and the data information of the energy consumption of the vehicle motor are input to determine the regression prediction function R of the BP neural network,

[0064]

[0065] Among them, r1 is the data information of the energy supplied by the hydrogen stack, r2 is the data information of the energy supplied by the power battery, r3 is the data information of the energy consumed by the vehicle motor, δ1, δ2 and δ3 are regression prediction factors, which predict the comprehensive energy loss of the vehicle and obtain the predicted data information of the comprehensive energy loss of the vehicle.

[0066] In this embodiment, the constraints of the regression prediction factors δ1, δ2 and δ3 are:

[0067] In this embodiment, the penalty coefficients β1, β2 and β3 are,

[0068]

[0069] Among them, y1 is the data information of the weight of the initialized BP neural network, and y2 is the data information of the bias of the initialized BP neural network.

[0070] Example 2: Based on the energy data processing system of a hydrogen-electric HIL model based on a power domain controller in Example 1, the present invention is further illustrated and described below.

[0071] like Figure 1 As shown, an energy data processing system of a hydrogen-electric HIL model based on a power domain controller includes a hydrogen stack energy supply module, a charging energy supply module, a motor consumption module, a comprehensive energy calculation module, a dynamic vehicle module, a working condition selection module, a speed parameter processing module and a slope parameter processing module.

[0072] The hydrogen stack energy supply module is used to obtain data information of the energy supplied by the hydrogen stack in real time;

[0073] The charging energy supply module is used to obtain data information on the energy supplied by the power battery in real time; the motor consumption module is used to obtain data information on the energy consumed by the motor of the whole vehicle in real time; the comprehensive energy calculation module is connected to the hydrogen stack energy supply module, the power supply module and the motor consumption module, and is used to predict the comprehensive energy loss of the whole vehicle based on the data information on the energy supplied by the hydrogen stack, the data information on the energy supplied by the power battery and the data information on the energy consumed by the motor of the whole vehicle, using a BP neural network regression prediction algorithm based on ridge regression optimization to obtain the predicted data information on the comprehensive energy loss of the whole vehicle; the dynamic vehicle module is connected to the comprehensive energy calculation module, and is used to construct a whole vehicle energy management model based on the vehicle dynamics model according to the predicted data information on the comprehensive energy loss of the whole vehicle and the usage data required by each actuator, so as to manage the energy of the whole vehicle and output the data information on the energy management of the whole vehicle.

[0074] In this embodiment, if Figure 3 As shown, the construction of the vehicle energy management model based on the vehicle dynamics model and the management of the energy of the vehicle include:

[0075] U1. Based on the predicted comprehensive energy loss data of the vehicle and the usage data required by each actuator, establish a relationship function P between energy loss and actuator usage,

[0076] Among them, z1 is the data information of the predicted comprehensive energy loss of the whole vehicle, z2 is the usage data information required to receive each actuator, γ1, γ2 and γ3 are relationship constant parameters, which characterize the relationship between the energy loss of the whole vehicle and the energy required by the actuator, and obtain the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator;

[0077] U2. Input the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator into the whole vehicle energy management model based on the vehicle dynamics model for training and learning, and determine the whole vehicle energy management function S,

[0078]

[0079] Among them, h is the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator, η1, η2 and η3 are the allocation factors of the whole vehicle energy management, and the trained whole vehicle energy management model based on the vehicle dynamics model is obtained;

[0080] U3. Based on the trained vehicle energy management model based on the vehicle dynamics model, the predicted data information of the comprehensive energy loss of the whole vehicle and the usage data required by each actuator are input, the energy of the whole vehicle is managed, and the data information of the energy management of the whole vehicle is output.

[0081] In this implementation, the constraint function f of the allocation factors η1, η2 and η3 of the vehicle energy management is,

[0082]

[0083] The value range of the constraint function f is (0,1).

[0084] In summary, the hydrogen-electric HIL model of the present invention not only increases the dynamics of the entire energy calculation, but also the operating condition selection module imports real vehicle data to improve data authenticity, categorizes parameters, and enhances the diversification of HIL model operation.

[0085] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. An energy data processing system of a hydrogen-electric HIL model based on a power domain controller, characterized in that: The system includes a hydrogen stack energy supply module, a charging energy supply module, a motor consumption module, a comprehensive energy calculation module, a dynamic vehicle module, a working condition selection module, a speed parameter processing module and a slope parameter processing module. The hydrogen stack energy supply module is used to obtain data information of the energy supplied by the hydrogen stack in real time; The charging energy supply module is used to obtain data information of energy supplied by the power battery in real time; the motor consumption module is used to obtain data information of energy consumed by the motor of the whole vehicle in real time; the comprehensive energy calculation module is connected to the hydrogen stack energy supply module, the power supply module and the motor consumption module, and is used to predict the comprehensive energy loss of the whole vehicle based on the data information of energy supplied by the hydrogen stack, the data information of energy supplied by the power battery and the data information of energy consumed by the motor of the whole vehicle, using the BP neural network regression prediction algorithm based on ridge regression optimization to obtain the predicted data information of the comprehensive energy loss of the whole vehicle; The dynamic vehicle module is connected to the comprehensive energy calculation module, and is used to build a whole vehicle energy management model based on the vehicle dynamics model according to the predicted data information of the comprehensive energy loss of the whole vehicle and the usage data required by each actuator, so as to manage the energy of the whole vehicle and output the data information of the energy management of the whole vehicle.

2. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 1 is characterized in that: The speed parameter processing module is used to process the speed change of the vehicle into data information of speed type, and the slope parameter processing module is used to process the slope change into data information of slope type.

3. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 2 is characterized in that: The working condition selection module is connected to the speed parameter processing module, the slope parameter processing module and the dynamic vehicle module, and is used to construct a working condition category selection function Q according to the data information of the energy management of the whole vehicle, the data information of the speed type, the data information of the slope type and the data information of the real vehicle of different cycle working conditions imported, Among them, x1 is the data information of the energy management of the whole vehicle, x2 is the data information of the speed type, x3 is the data information of the slope type, x4 is the data information of the actual vehicle under different imported cycle conditions, α1, α2 and α3 are the decision factors of the working condition category, which characterize the category selection of the working condition and obtain the data information of the working condition selection category.

4. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 3 is characterized in that: The decision factors α1, α2 and α3 of the working condition category are: Among them, x1 is the data information of the energy management of the whole vehicle, x2 is the data information of the speed type, x3 is the data information of the slope type, and x4 is the data information of the actual vehicle under different imported cycle conditions.

5. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 3 is characterized by: The working condition selection module is connected to the working condition display module of the host computer, and the working condition display module of the host computer is used to display data information of the selected category of the working condition in real time.

6. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 1 is characterized in that: The method of using the BP neural network regression prediction algorithm based on ridge regression optimization to predict the comprehensive energy loss of the whole vehicle includes: L1. Inputting the energy supply data information of the hydrogen stack, the energy supply data information of the power battery and the energy consumption data information of the vehicle motor into the BP neural network, initializing the weight and bias of the BP neural network, and obtaining the weight and bias data information of the initialized BP neural network; L2. Based on the weight and bias data information of the initialized BP neural network, a ridge regression optimization function W is established. Among them, y1 is the data information of the weight of the initialized BP neural network, y2 is the data information of the bias of the initialized BP neural network, β1, β2 and β3 are penalty coefficients, and the weight and bias of the BP neural network are optimized to obtain the optimized BP neural network; L3. Based on the optimized BP neural network, the data information of the energy supply of the hydrogen stack, the data information of the energy supply of the power battery and the data information of the energy consumption of the vehicle motor are input to determine the regression prediction function R of the BP neural network, Among them, r1 is the data information of the energy supplied by the hydrogen stack, r2 is the data information of the energy supplied by the power battery, r3 is the data information of the energy consumed by the vehicle motor, δ1, δ2 and δ3 are regression prediction factors, which predict the comprehensive energy loss of the vehicle and obtain the predicted data information of the comprehensive energy loss of the vehicle.

7. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 6 is characterized by: The constraints of the regression prediction factors δ1, δ2 and δ3 are:

8. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 6 is characterized by: The penalty coefficients β1, β2 and β3 are, Among them, y1 is the data information of the weight of the initialized BP neural network, and y2 is the data information of the bias of the initialized BP neural network.

9. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 1 is characterized in that: The construction of the vehicle energy management model based on the vehicle dynamics model to manage the energy of the vehicle includes: U1. Based on the predicted comprehensive energy loss data of the vehicle and the usage data required by each actuator, establish a relationship function P between energy loss and actuator usage, Among them, z1 is the data information of the predicted comprehensive energy loss of the whole vehicle, z2 is the usage data information required to receive each actuator, γ1, γ2 and γ3 are relationship constant parameters, which characterize the relationship between the energy loss of the whole vehicle and the energy required by the actuator, and obtain the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator; U2. Input the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator into the whole vehicle energy management model based on the vehicle dynamics model for training and learning, and determine the whole vehicle energy management function S, Among them, h is the data information of the relationship between the energy loss of the whole vehicle and the energy required by the actuator, η1, η2 and η3 are the allocation factors of the whole vehicle energy management, and the trained whole vehicle energy management model based on the vehicle dynamics model is obtained; U3. Based on the trained vehicle energy management model based on the vehicle dynamics model, the predicted data information of the comprehensive energy loss of the whole vehicle and the usage data required by each actuator are input, the energy of the whole vehicle is managed, and the data information of the energy management of the whole vehicle is output.

10. The energy data processing system of the hydrogen-electric HIL model based on the power domain controller according to claim 9 is characterized in that: The constraint function f of the allocation factors η1, η2 and η3 of the vehicle energy management is, The value range of the constraint function f is (0,1).

Images