CN111079349B - Energy real-time optimization method for lithium battery and super capacitor composite power supply system - Google Patents
Energy real-time optimization method for lithium battery and super capacitor composite power supply system Download PDFInfo
- Publication number
- CN111079349B CN111079349B CN201911384077.2A CN201911384077A CN111079349B CN 111079349 B CN111079349 B CN 111079349B CN 201911384077 A CN201911384077 A CN 201911384077A CN 111079349 B CN111079349 B CN 111079349B
- Authority
- CN
- China
- Prior art keywords
- lithium battery
- input range
- fuzzy
- quantized
- supply system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L50/00—Electric propulsion with power supplied within the vehicle
- B60L50/40—Electric propulsion with power supplied within the vehicle using propulsion power supplied by capacitors
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L50/00—Electric propulsion with power supplied within the vehicle
- B60L50/50—Electric propulsion with power supplied within the vehicle using propulsion power supplied by batteries or fuel cells
- B60L50/60—Electric propulsion with power supplied within the vehicle using propulsion power supplied by batteries or fuel cells using power supplied by batteries
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
- B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
- B60L58/12—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries responding to state of charge [SoC]
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
- B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
- B60L58/16—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries responding to battery ageing, e.g. to the number of charging cycles or the state of health [SoH]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Power Engineering (AREA)
- Theoretical Computer Science (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Transportation (AREA)
- Sustainable Energy (AREA)
- Sustainable Development (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention provides a real-time energy optimization method for a lithium battery and super capacitor composite power supply system, which is characterized in that health factors of a lithium battery are taken into consideration of fuzzy energy management rules of the composite power supply system, the power of the lithium battery and the power of a super capacitor are reasonably distributed, the advantages of the super capacitor and the lithium battery are exerted, and the service life of the lithium battery is prolonged; the fuzzy variable adopts a triangular membership function, the fuzzy variable is used for measuring the average value of the input range of the fuzzy variable, the datamation of a fuzzy rule is realized, and an input data sample and an output data sample are obtained; learning the fuzzy rule after datamation by using the learning ability of the neural network to obtain a fuzzy energy management module based on the neural network; on the basis, the root mean square of the output current of the lithium battery is used as a performance function, the weight and the threshold of the neural network are adjusted by using a BP algorithm, the energy management rule of the composite power supply system is optimized, and the service life of the lithium battery is prolonged while the required power of the composite power supply system is met.
Description
Technical Field
The invention relates to the field of energy management of a composite power supply system of an electric automobile, in particular to a real-time energy optimization method for a lithium battery and super capacitor composite power supply system.
Background
The electric automobile has the advantages of environmental protection and development potential. At present, the electric automobile is still in a growth stage, and the market share of the electric automobile is very low. The main reason is that the performance and service life of the lithium battery are difficult to meet the requirements of the electric automobile.
The energy storage and charge-discharge processes of the lithium battery are electrochemical processes, and high-rate discharge reduces the cycle life of the lithium battery, so that the capacity of the lithium battery is reduced. The larger the discharge rate of the lithium battery is, the larger the influence on the service life of the lithium battery is.
Because the electric automobile has a high power peak-to-average ratio when running on a road, the capacity loss and the service life of the electric automobile are influenced by the high-rate discharge of the lithium battery during acceleration and climbing.
The super capacitor with high power density and strong charging and discharging capacity is used as an auxiliary energy source, and is an effective way for prolonging the service life of the lithium battery, improving the acceleration performance of the electric automobile and the like. The service life of the lithium battery is prolonged, so that the reliability of the electric automobile is improved, the cost of the electric automobile is reduced, and the competitiveness of the electric automobile is improved.
The energy management system of the composite power supply system is a key for reasonably distributing the power of the lithium battery and the power of the super capacitor, exerting the advantages of the super capacitor and the lithium battery and prolonging the service life of the lithium battery.
In practical application, the health condition of the lithium battery gradually declines, so that the output power of the lithium battery is suitable for the health condition of the lithium battery. However, in the conventional fuzzy energy management rule, the health condition of the lithium battery pack is not fully considered.
At present, the optimization of the energy management rule of the composite power supply system is performed by adopting methods such as a genetic algorithm and the like for off-line optimization, and the on-line optimization method is less.
Disclosure of Invention
In view of the above, the invention provides a method for optimizing the energy of a lithium battery and super capacitor composite power supply system in real time, which can meet the power demand of the composite power supply system and prolong the service life of the lithium battery according to the health state of the lithium battery.
The technical scheme of the invention is realized as follows: the invention provides a method for optimizing the energy of a lithium battery and super capacitor composite power supply system in real time, which comprises the following steps,
s1, inputting variable lithium battery health factor lambda bt,soh And the required power P of the hybrid power system hes,r And super capacitor state of charge SOC sc (ii) a Quantizing and fuzzifying the input variables, and converting the quantized and fuzzified variables into fuzzy quantities;
s2, obtaining the required power P of the lithium battery pack based on the fuzzy energy management rule bt,r The fuzzy quantity is defuzzified and quantized to obtain the required power P of the lithium battery pack bt,r ；
S3, the required power P of the lithium battery bt,r Transmitting to the lithium battery pack module to obtain the output power P of the lithium battery pack bt,a 。
On the basis of the above technical solution, preferably, the method further comprises the following steps,
s4, digitizing the fuzzy energy management rule to obtain an input and output data sample;
and S5, performing off-line learning on the data sample by using the BP neural network to obtain the neural network energy management module memorizing the fuzzy energy management rule.
Further preferably, in step S1,
input variable lithium battery health factor lambda bt,soh The quantized input range is [0,1 ]]The fuzzy set is (L, M, H), H, M and L represent high, middle and low respectively; when lambda is bt,soh The quantized input range is 0,0.33]Then, corresponding to the blur amount L; when lambda is bt,soh The quantized input range is [0.33,0.66 ]]When the value is zero, corresponding to the fuzzy quantity M; when lambda is bt,soh The quantized input range is 0.66,1]When so, corresponding to the fuzzy amount H;
input variable composite power supply system required power P hes,r The quantized input range is [ -1,1]The fuzzy set is (N, L, M, H), H, M, L and N represent high, middle, low and negative respectively; when P is present hes,r The quantized input range is [ -1,0 []Then, corresponding to the fuzzy quantity N; when P is present hes,r The quantized input range is 0,0.33]Then, corresponding to the blur amount L; when P is present hes,r The quantized input range is [0.33,0.66 ]]Then, corresponding to the fuzzy quantity M; when P is present hes,r The quantized input range is 0.66,1]When so, corresponding to the fuzzy amount H;
input variable super capacitor state of charge SOC sc The quantized input range is [0,1 ]]The fuzzy sets are (L, M, H, M and L represent high, middle and low respectively; when the SOC is sc The quantized input range is 0,0.57]Then, corresponding to the blur amount L; when SOC is reached sc The input range after quantization is [0.57,0.81 ]]When the value is zero, corresponding to the fuzzy quantity M; when SOC is reached sc The quantized input range is [0.81,1 ]]When so, corresponding to the fuzzy amount H;
input variable lithium battery pack required power P bt,r The quantized input range is [ -0.3,1]The fuzzy set is (N, L, M, H), H, M, L and N represent high, medium, low and negative respectively; when P is present bt,r The quantized input range is [ -0.3,0]Then, corresponding to the fuzzy quantity N; when P is present bt,r The quantized input range is [0,0.33 ]]Then, corresponding to the blur amount L; when P is present bt,r The quantized input range is [0.33,0.66 ]]When the value is zero, corresponding to the fuzzy quantity M; when P is present bt,r The quantized input range is[0.66，1]When so, corresponding to the fuzzy amount H;
in step S2, the required power P of the lithium battery pack is obtained according to the following fuzzy energy management rule bt,r Amount of blur of (2):
when the health factor lambda of the lithium battery bt,soh When the amount of blurring is set to H,
when the health factor lambda of the lithium battery bt,soh When the amount of blurring is M, the amount of blurring is,
when the health factor lambda of the lithium battery bt,soh When the amount of blurring is L,
still preferably, in step S4, the fuzzy energy management rule is digitalized in the following manner:
when lambda is bt,soh The quantized input range is [0,0.33 ]]Corresponding to the fuzzy quantity L, taking a median value of an input range to be 0.165 during datamation; when lambda is bt,soh The quantized input range is [0.33,0.66 ]]Corresponding to the fuzzy quantity M, taking a median value of an input range to be 0.485 during datamation; when lambda is bt,soh The quantized input range is [0.66, 1']Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation;
when P is present hes,r The quantized input range is [ -1,0 ]]Corresponding to the fuzzy quantity N, taking a median value of-0.5 in an input range during datamation; when P is present hes,r The quantized input range is 0,0.33]When the fuzzy quantity L is needed, the median value in the input range is 0.165 corresponding to the fuzzy quantity L; when P is present hes,r The quantized input range is [0.33,0.66 ]]Corresponding to the fuzzy quantity M, taking a median value of an input range to be 0.485 during datamation; when P is present hes,r Quantized input rangeIs [0.66,1 ]]Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation;
when SOC is reached sc The quantized input range is [0,0.57 ]]Corresponding to the fuzzy quantity L, taking E when the data is formed sc Inputting a square root value 0.41 of a median value 0.165 in the range; when SOC is reached sc The input range after quantization is [0.57,0.81 ]]When the data is converted into data, corresponding to the fuzzy quantity M, the value E is taken sc Inputting a square root value of 0.70 of a median value of 0.495 in the range; when SOC is reached sc The quantized input range is [0.81,1 ]]When the data is converted into data, corresponding to the fuzzy quantity H, the fuzzy quantity E is taken sc Inputting a quadratic root value 0.91 of a median value 0.825 in the range;
when P is bt,r The quantized input range is [ -0.3,0]Corresponding to the fuzzy quantity N, taking a median value of-0.15 in an input range during datamation; when P is present bt,r The quantized input range is 0,0.33]Corresponding to the fuzzy quantity L, taking a median value of an input range to be 0.165 during datamation; when P is present bt,r The quantized input range is [0.33,0.66 ]]Corresponding to the fuzzy quantity M, taking a median value of an input range to be 0.485 during datamation; when P is present bt,r The quantized input range is 0.66,1]Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation;
the input data sample of the neural network is a lithium battery health factor lambda bt,soh Required power P of composite power supply system hes,r And super capacitor state of charge SOC sc The output data sample of the neural network is the required power P of the lithium battery bt,r 。
Further preferably, the method further comprises the following steps,
s6, selecting the root mean square of the output current of the lithium battery as a performance function, utilizing the self-learning capability of the neural network, adopting a BP algorithm to adjust the weight and the threshold of the neural network on line, and optimizing the fuzzy energy management rule in real time to enable the overall performance function to be minimum.
Still more preferably, in step S5,
the quantization module converts the actual value of the input variable into the input range of the neural network, and the weight coefficient from the input layer to the hidden layer is w at the moment of k (2) ij (k)(i＝1,2…9,j＝1,2,3) The threshold is b (2) i (k) (i =1,2 \82309; 9), hidden-to-output-layer weight coefficient w (3) li (k) (i =1,2 \8230; 9,l = 1), and the threshold value is b (3) l (k)(l＝1)；
In the step S6, the root mean square of the output current of the lithium battery pack representing the service life of the lithium battery is taken as a performance function J (k) of the neural network,
I bt,a (k) Represents the output current of the lithium battery pack at the moment k, and N represents the output current I of the lithium battery pack at the moment k bt,a The total number of sampling values;
when the input variable is transmitted in the forward direction by using a BP learning algorithm, the input variable is transmitted from an input layer to an output layer through a hidden layer; when reversely propagating, transmitting the performance function J (k) from the output layer to the input layer through the hidden layer, and adjusting the weight and the threshold of the neural network on line according to a gradient descent method; searching the negative gradient direction of the neural network according to J (k), adding an inertia term which enables the search to be rapidly converged, and optimizing an energy management strategy, wherein a weight value and threshold value adjusting formula is as follows:
where η is the learning rate and α is the inertia coefficient.
Further preferably, the method further comprises a step S7 of evaluating the optimization result, and when the optimization result meets the requirement, the optimization is finished; and when the requirements cannot be met, modifying the fuzzy energy management rule, and optimizing according to the steps S1-S6.
On the basis of the technical scheme, preferably, the energy real-time optimization method is completed in ADVISOR software based on MATLAB.
Compared with the prior art, the method for optimizing the energy of the lithium battery and super capacitor composite power supply system in real time has the following beneficial effects:
(1) The health factors of the lithium battery are brought into the fuzzy energy management rule consideration range of the composite power system, the power of the lithium battery and the power of the super capacitor are reasonably distributed, the advantages of the super capacitor and the lithium battery are exerted, and the service life of the lithium battery is prolonged;
(2) The fuzzy variable adopts a triangular membership function, the fuzzy variable is used for measuring the average value of the input range of the fuzzy variable, the datamation of a fuzzy rule is realized, and an input data sample and an output data sample are obtained;
(3) Learning the fuzzy rule after the datamation by using the learning ability of the neural network to obtain a fuzzy energy management module based on the neural network;
(4) On the basis, the root mean square of the output current of the lithium battery is used as a performance function, the weight and the threshold of the neural network are adjusted by using a BP algorithm, the energy management rule of the composite power supply system is optimized, and the service life of the lithium battery is prolonged while the required power of the composite power supply system is met.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 shows a real-time energy optimization process of a lithium battery and super capacitor hybrid power system;
FIG. 2 is a diagram of a fuzzy energy management structure of a hybrid power system based on ADVISOR software;
FIG. 3 is a MATLAB based composite power system fuzzy energy management module;
FIG. 4 is a MATLAB-based composite power system fuzzy energy management rule;
FIG. 5 MATLAB-based lithium battery health factor λ bt,soh A membership function graph;
FIG. 6 MATLAB-based composite power supply system required power P hes,r A membership function graph;
FIG. 7 super-capacitor state-of-charge SOC based on MATLAB sc A membership function graph;
FIG. 8 lithium battery pack demand power P based on MATLAB bt,r A membership function graph;
FIG. 9 is a diagram of a BP neural network architecture;
FIG. 10 is a diagram of a neural network real-time optimization energy management structure based on ADVIOR software.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments of the present invention, belong to the protection scope of the present invention.
The ADVISOR software is source code open electric vehicle simulation software which is developed based on MATLAB software and can be used for carrying out forward simulation and backward simulation, and is suitable for energy management evaluation and optimization of electric vehicles. The energy real-time optimization method of the lithium battery and super capacitor composite power supply system is completed in ADVISOR2002 software based on MATLAB6.5 environment.
The energy real-time optimization process of the lithium battery and super capacitor composite power supply system is shown in fig. 1, and comprises the following steps:
1. formulating fuzzy energy management rules
According to the characteristics of small energy density of lithium batteries and large power density and small energy density of super capacitors, the fuzzy energy management rule of the composite power supply system considering the health condition of the lithium batteries is as follows:
TABLE 1 (a) lithium cell health factor λ bt,soh When the amount of blurring is H, the amount of blurring is,
TABLE 1 (b) lithium cell health factor λ bt,soh When the amount of blurring is M, the image blur correction is performed,
TABLE 1 (c) lithium cell health factor lambda bt,soh When the amount of blurring is L,
wherein λ is bt,soh Represents the input variable lithium battery health factor, P hes,r Representing the demanded power, SOC, of the input variable hybrid power system sc Representing the input variable super capacitor state of charge; p bt,r Representing the output variable lithium battery pack required power.
As shown in the table above, as the state of health of the lithium battery decreases, the required power of the lithium battery decreases accordingly. The required power of the lithium battery pack is properly adjusted according to the health condition factor of the lithium battery, so that the over-current charging and discharging of the lithium battery are avoided, and the service life of the lithium battery is prolonged.
Thirty-six fuzzy rules are shared in tables 1 (a), 1 (b) and 1 (c). In Table 1 (a), one of the fuzzy energy management rules is described as the lithium battery health factor λ bt,soh H (high), the composite power system requires power P hes,r Is H (high), state of charge SOC of super capacitor sc When M is middle, the lithium battery pack needs power P bt,r Is H (high).
In Table 1 (b), one of the fuzzy energy management rules is describedAs mentioned above, the health factor lambda of lithium battery bt,soh For M (middle), the compound power system requires power P hes,r Is H (high), state of charge SOC of super capacitor sc When M (middle), the lithium battery pack requires power P bt,r Is M (middle).
In Table 1 (C), one of the fuzzy energy management rules is described as the lithium battery health factor λ bt,soh Is L (low), the composite power supply system requires power P hes,r H (high), state of charge SOC of the supercapacitor sc When M is middle, the lithium battery pack needs power P bt,r L (low).
In the three fuzzy energy management rules, the required power P of the lithium battery pack bt,r The health condition of the lithium battery is considered, and the service life of the lithium battery is prolonged.
The structure diagram of the hybrid power system fuzzy energy management module based on the ADVISOR software is shown in FIG. 2. P hes,r Power demand for hybrid power supply system, P hes,a For outputting power, P, to the combined power supply system sc,r Power requirement for super capacitor, P sc,a For output of power, P, from the super-capacitor bt,r Power requirement for lithium battery, P bt,a And outputting power for the lithium battery.
In the power distribution module, the super capacitor bank requires power P sc,r The following formula.
P sc,r ＝P hes,r -P bt,a (1)
Output power P of composite power supply system hes,a The following formula.
P hes,a ＝P bt,a +P sc,a (2)
A MATLAB-based fuzzy energy management module is shown in fig. 3. Fuzzy energy management module inputs variable lithium battery health factor lambda bt,soh Required power P of composite power supply system hes,r And super capacitor state of charge SOC sc The input variables are quantized and fuzzified to convert into fuzzy quantities. Obtaining the required power P of the lithium battery by inquiring the fuzzy energy management rules of the table 1 (a), the table 1 (b) and the table 1 (c) bt,r The fuzzy quantity is defuzzified and quantized to obtain the required work of the lithium batteryRate P bt,r . Lithium battery demand power P bt,r Transmitting to the lithium battery pack module to obtain the output power P of the lithium battery pack bt,a 。
The MATLAB-based fuzzy energy management rules are obtained from the fuzzy energy management rules of tables 1 (a), 1 (b), and 1 (c) and are shown in fig. 4.
Required power P of super capacitor bank sc,r The required power P of the super capacitor bank is obtained by the formula (1) sc,r Transmitting to the super capacitor bank module to obtain the output power P of the super capacitor bank sc,a 。
2. Datamation fuzzy energy management rule
Lithium battery health factor lambda based on MATLAB bt,soh The graph of membership functions is shown in FIG. 5, using triangular membership functions. The lithium battery health condition evaluation module obtains a lithium battery health condition factor lambda according to the voltage, the current, the running time and other parameters of the lithium battery bt,soh . Input variable lithium battery health factor lambda bt,soh The quantized input range is [0,1 ]]The fuzzy set is (L, M, H), H, M and L represent high, middle and low respectively; when lambda is bt,soh The quantized input range is [0,0.33 ]]Corresponding to the fuzzy quantity L, taking a median value of an input range to be 0.165 during datamation; when lambda is bt,soh The quantized input range is [0.33,0.66 ]]Corresponding to the fuzzy quantity M, taking a median value of an input range to be 0.485 during datamation; when lambda is bt,soh The quantized input range is [0.66, 1']Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation; the membership functions of L (low), M (medium) and H (high) are respectively,
y＝trimf(x,[0,0.165,0.33])
y＝trimf(x,[0.33,0.485,0.66])
y＝trimf(x,[0.66,0.825,1])
MATLAB-based composite power supply system required power P hes,r The graph of membership functions is shown in FIG. 6, using triangular membership functions. Input variable composite power supply system required power P hes,r The quantized input range is [ -1,1]The fuzzy set is (N, L, M, H), H, M, L and N represent high, middle, low and negative respectively; when P is hes,r The quantized input range is [ -1,0 ]]Corresponding to the fuzzy quantity N, taking a median value of-0.5 in an input range during datamation; when P is hes,r The quantized input range is [0,0.33 ]]When the fuzzy quantity L is needed, the median value in the input range is 0.165 corresponding to the fuzzy quantity L; when P is present hes,r The quantized input range is [0.33,0.66 ]]Corresponding to the fuzzy quantity M, taking a median value of an input range to be 0.485 during datamation; when P is present hes,r The quantized input range is 0.66,1]Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation; the membership degree functions of N (negative), L (low), M (medium) and H (high) are respectively,
y＝trimf(x,[-1,-0.5,0])
y＝trimf(x,[0,0.165,0.33])
y＝trimf(x,[0.33,0.485,0.66])
y＝trimf(x,[0.66,0.825,1])
super capacitor state of charge SOC based on MATLAB sc The membership function graph is shown in fig. 7, using triangular membership functions. Input variable super capacitor state of charge SOC sc The quantized input range is [0,1 ]]The fuzzy sets are (L, M, H, M and L represent high, middle and low respectively; energy E stored due to super capacitor sc And state of charge SOC of super capacitor sc Is proportional to the square of when the SOC is sc The quantized input range is [0,0.57 ]]
Corresponding to the fuzzy quantity L, taking E when the data is formed sc Inputting a quadratic root value of 0.41 of a median value of 0.165 in the range; when SOC is reached sc The quantized input range is [0.57,0.81 ]]When the data is converted into data, corresponding to the fuzzy quantity M, the value E is taken sc Inputting a quadratic root value 0.70 of a range median value 0.495; when SOC is reached sc The quantized input range is 0.81,1]When the data is converted into data, corresponding to the fuzzy quantity H, the fuzzy quantity E is taken sc Inputting a square root value 0.91 of a median value 0.825 in the range; the membership functions of L (low), M (medium), H (high) are respectively,
y＝trimf(x,[0,0.41,0.57])
y＝trimf(x,[0.57,0.7,0.81])
y＝trimf(x,[0.81,0.91,1])
lithium battery pack required power P based on MATLAB bt,r The graph of membership functions is shown in FIG. 8, using triangular membership functions. Input variable lithium battery pack required power P bt,r The quantized input range is [ -0.3,1]The fuzzy set is (N, L, M, H), H, M, L and N represent high, middle, low and negative respectively; when P is present bt,r The quantized input range is [ -0.3,0]Corresponding to the fuzzy quantity N, taking a median value of-0.15 in an input range during datamation; when P is present bt,r The quantized input range is 0,0.33]Corresponding to the fuzzy quantity L, taking a median value of an input range to be 0.165 during datamation; when P is present bt,r The quantized input range is [0.33,0.66 ]]Corresponding to the fuzzy quantity M, taking a median value of an input range to be 0.485 during datamation; when P is present bt,r The quantized input range is [0.66, 1']Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation; the membership functions of N (negative), L (low), M (medium) and H (high) are respectively,
y＝trimf(x,[-0.3,-0.15,0])
y＝trimf(x,[0,0.165,0.33])
y＝trimf(x,[0.33,0.485,0.66])
y＝trimf(x,[0.66,0.825,1])
3. input and output data samples
Thirty-six fuzzy control rules in the tables 1 (a), 1 (b) and 1 (c), after data processing, the input data sample of the neural network is the lithium battery health factor lambda bt,soh Required power P of composite power supply system hes,r And super capacitor state of charge SOC sc The output data sample of the neural network is the required power P of the lithium battery bt,r Input data samples P for neural network training as shown in table 2 (a) and output data samples T as shown in table 2 (b) are obtained.
TABLE 2 (a) input samples P of neural networks
P＝[0.165 -0.5 0.41；0.165 -0.5 0.70；0.165 -0.5 0.91；0.165 0.165 0.41；0.165 0.165 0.70；0.165 0.165 0.91；0.165 0.495 0.41；0.165 0.495 0.70；0.165 0.495 0.91；0.165 0.825 0.41；0.165 0.825 0.70；0.165 0.825 0.91；0.495 -0.5 0.41；0.495 -0.5 0.70；0.495 -0.5 0.91；0.495 0.165 0.41；0.495 0.165 0.70；0.495 0.165 0.91；0.495 0.495 0.41；0.495 0.495 0.70；0.495 0.495 0.91；0.495 0.825 0.41；0.495 0.825 0.70；0.495 0.825 0.91；0.825 -0.5 0.41；0.825 -0.5 0.70；0.825 -0.5 0.91；0.825 0.165 0.41；0.825 0.165 0.70；0.825 0.165 0.91；0.825 0.495 0.41；0.825 0.495 0.70；0.825 0.495 0.91；0.825 0.825 0.41；0.825 0.825 0.70；0.825 0.825 0.91；]＇
TABLE 2 (a) output samples T of neural networks
T = [0.165; -0.15; -0.15;0.495;0.165;0.165;0.825;0.495;0.165;0.825;0.825;0.495;0.165; -0.15; -0.15;0.495; -0.15; -0.15;0.495;0.495; -0.15;0.495;0.495;0.495;0.165; -0.15; -0.15;0.165;0.165;0.165;0.495;0.165;0.165;0.495;0.165;0.165; ' 4. Neural network off-line learning
The structure of the BP neural network is shown in FIG. 9. The input layer is provided with three neuron nodes, the output layer is provided with one neuron node, the number of the neuron nodes of the hidden layer is generally obtained according to an empirical formula and experiments, and 9 nodes can be selected. The input data sample of the neural network is a lithium battery health factor lambda bt,soh And the required power P of the hybrid power system hes,r And super capacitor state of charge SOC sc The output data sample of the neural network is the required power P of the lithium battery bt,r . The quantization module converts the actual value of the input variable into the input range of the neural network, and the weight coefficient from the input layer to the hidden layer is w (2) ij (k) (j =1,2,3, i =1,2 \ 82309; 9), and the threshold value is b (2) i (k) (i =1,2 \82309; 9), the hidden-to-output layer weight coefficient is w (3) li (k) (i =1,2 \8230; 9,l = 1), and the threshold value is b (3) l (k) (l = 1). Based on MATLAB neural network toolbox, using BP learning algorithm to apply input data sample P and output data sample T to neural network off-line training, selecting neural network with minimum error between output quantity of neural network and output data sample T, and realizing mapping between input data sample P and output data sample T by the neural networkAnd the neural network memorizes fuzzy rules and realizes the energy management of the composite power supply system based on the neural network.
5. Energy management online optimization of composite power supply system based on neural network
The energy management structure of the neural network-based hybrid power system is shown in fig. 10.
The root mean square of the output current of the lithium battery pack, which characterizes the life of the lithium battery, is taken as a neural network performance function J (k), which can be expressed as,
I bt,a (k) Represents the output current of the lithium battery pack at the moment k, and N represents the output current I of the lithium battery pack at the moment k bt,a Total number of sample values.
The neural network adopts the trained neural network. By utilizing a BP learning algorithm, when the forward propagation is carried out, an input variable is transmitted to an output layer from an input layer through a hidden layer; during reverse propagation, a performance function J (k) is transmitted from an output layer to an input layer through a hidden layer, and the weight and the threshold of the neural network are adjusted on line according to a gradient descent method; searching the negative gradient direction of the neural network according to J (k), adding an inertia term which enables the search to be rapidly converged, and optimizing an energy management strategy, wherein a weight value and threshold value adjusting formula is as follows:
where η is the learning rate and α is the inertia coefficient.
6. Evaluation of
And evaluating the optimization result, and finishing the optimization when the optimization result meets the requirement. And when the requirements cannot be met, modifying fuzzy energy management rules, and then optimizing according to the flow.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (8)
1. A real-time energy optimization method for a lithium battery and super capacitor composite power supply system is characterized by comprising the following steps: comprises the following steps of (a) carrying out,
s1, inputting variable lithium battery health factor lambda bt,soh And the required power P of the hybrid power system hes,r And super capacitor state of charge SOC sc (ii) a Quantizing and fuzzifying the input variables, and converting the quantized and fuzzified variables into fuzzy quantities;
s2, obtaining the required power P of the lithium battery pack based on the fuzzy energy management rule bt,r The fuzzy quantity is defuzzified and quantized to obtain the required power P of the lithium battery pack bt,r (ii) a In step S2, the required power P of the lithium battery pack is obtained according to the following fuzzy energy management rule bt,r Amount of blur of (2):
when the health factor lambda of the lithium battery bt,soh When the amount of blurring is H, the amount of blurring is,
when the health factor lambda of the lithium battery bt,soh When the amount of blurring is M, the image blur correction is performed,
when the health factor lambda of the lithium battery bt,soh When the amount of blurring is L,
H. m, L and N, representing high, medium, low and negative, respectively;
s3, the required power P of the lithium battery bt,r Transmitting to the lithium battery pack module to obtain the output power P of the lithium battery pack bt,a 。
2. The energy real-time optimization method for the lithium battery and supercapacitor composite power supply system according to claim 1, characterized in that: the method also comprises the following steps of,
s4, digitizing the fuzzy energy management rule to obtain an input and output data sample;
and S5, performing off-line learning on the data sample by using the BP neural network to obtain the neural network energy management module memorizing the fuzzy energy management rule.
3. The energy real-time optimization method for the lithium battery and supercapacitor composite power supply system according to claim 2, characterized in that:
in the step S1, the first step is carried out,
input variable lithium battery health factor lambda bt,soh The quantized input range is [0,1 ]]The fuzzy sets are (L, M, H, M and L represent high, middle and low respectively; when lambda is bt,soh The quantized input range is [0, 0.33), corresponding to the fuzzy quantity L; when lambda is bt,soh The quantized input range is [0.33, 0.66), corresponding to the blur amount M; when lambda is bt,soh The quantized input range is [0.66, 1']When so, corresponding to the fuzzy amount H;
input variable composite power supply system required power P hes,r The quantized input range is [ -1,1]The fuzzy set is (N, L, M, H), H, M, L and N represent high, middle, low and negative respectively; when P is present hes,r When the quantized input range is [ -1, 0), corresponding to the fuzzy quantity N; when P is present hes,r QuantizationThe later input range is [0, 0.33), corresponding to the fuzzy quantity L; when P is present hes,r The quantized input range is [0.33, 0.66), corresponding to the blur amount M; when P is present hes,r The quantized input range is [0.66, 1']When so, corresponding to the fuzzy amount H;
input variable super capacitor state of charge SOC sc The quantized input range is [0,1 ]]The fuzzy set is (L, M, H), H, M and L represent high, middle and low respectively; when SOC is reached sc The quantized input range is [0, 0.57), corresponding to the blur amount L; when the SOC is sc When the quantized input range is [0.57, 0.81), corresponding to the fuzzy quantity M; when the SOC is sc The quantized input range is [0.81,1 ]]When so, corresponding to the fuzzy amount H;
input variable lithium battery pack required power P bt,r The quantized input range is [ -0.3,1]The fuzzy set is (N, L, M, H), H, M, L and N represent high, middle, low and negative respectively; when P is present bt,r When the quantized input range is [ -0.3, 0), the fuzzy quantity N is corresponded to; when P is bt,r The quantized input range is [0, 0.33), corresponding to the fuzzy quantity L; when P is present bt,r The quantized input range is [0.33, 0.66), corresponding to the blur amount M; when P is present bt,r The quantized input range is [0.66, 1']And, in time, corresponds to the blur amount H.
4. The energy real-time optimization method for the lithium battery and supercapacitor composite power supply system according to claim 3, characterized in that:
in step S4, the fuzzy energy management rule is digitalized in the following manner:
when lambda is bt,soh When the quantized input range is [0, 0.33), corresponding to the fuzzy quantity L, and taking the median value of the input range to be 0.165 during the digitization; when lambda is bt,soh When the quantized input range is [0.33, 0.66), corresponding to the fuzzy quantity M, and taking the median value of the input range to be 0.485 during the digitization; when lambda is bt,soh The quantized input range is [0.66, 1']Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation;
when P is present hes,r When the quantized input range is [ -1, 0), corresponding to the fuzzy quantity N, dataTaking a median value of-0.5 in an input range during the formation; when P is present hes,r When the quantized input range is [0, 0.33), corresponding to the fuzzy quantity L, taking the median value of the input range to be 0.165; when P is hes,r When the quantized input range is [0.33, 0.66), corresponding to the fuzzy quantity M, and taking the median value of the input range to be 0.485 during the digitization; when P is present hes,r The quantized input range is 0.66,1]Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation;
when SOC is reached sc When the input range after quantization is [0, 0.57), corresponding to the fuzzy quantity L, the sampling E is taken during the data transformation sc Inputting a quadratic root value of 0.41 of a median value of 0.165 in the range; when SOC is reached sc When the input range after quantization is [0.57, 0.81), corresponding to the fuzzy quantity M, taking E when data is converted sc Inputting a square root value of 0.70 of a median value of 0.495 in the range; when SOC is reached sc The quantized input range is 0.81,1]When the data is converted into data, corresponding to the fuzzy quantity H, the fuzzy quantity E is taken sc Inputting a quadratic root value 0.91 of a median value 0.825 in the range;
when P is present bt,r When the quantized input range is [ -0.3, 0), corresponding to the fuzzy quantity N, and taking the median value of the input range to-0.15 during the digitization; when P is present bt,r When the quantized input range is [0, 0.33), corresponding to the fuzzy quantity L, and taking the median value of the input range to be 0.165 during the digitization; when P is present bt,r When the quantized input range is [0.33, 0.66), corresponding to the fuzzy quantity M, and taking the median value of the input range to be 0.485 during the digitization; when P is present bt,r The quantized input range is [0.66, 1']Corresponding to the fuzzy quantity H, taking a median value of an input range to be 0.825 during datamation;
the input data sample of the neural network is a lithium battery health factor lambda bt,soh Required power P of composite power supply system hes,r And super capacitor state of charge SOC sc The output data sample of the neural network is the required power P of the lithium battery bt,r 。
5. The energy real-time optimization method for the lithium battery and supercapacitor composite power supply system according to claim 2, characterized in that: the method also comprises the following steps of,
s6, selecting the root mean square of the output current of the lithium battery as a performance function, utilizing the self-learning capability of the neural network, adopting a BP algorithm to adjust the weight and the threshold of the neural network on line, and optimizing the fuzzy energy management rule in real time to enable the overall performance function to be minimum.
6. The energy real-time optimization method for the lithium battery and supercapacitor composite power supply system according to claim 5, characterized in that: in a step S5, the first step is executed,
the quantization module converts the actual value of the input variable into the input range of the neural network, and the weight coefficient from the input layer to the hidden layer is w at the moment of k (2) ij (k) (i =1,2 82309; 9,j =1,2,3) and the threshold value is b (2) i (k) (i =1,2 \82309; 9), hidden-to-output-layer weight coefficient w (3) li (k) (i =1,2 \8230; 9,l = 1), and the threshold value is b (3) l (k)(l＝1)；
In the step S6, the root mean square of the output current of the lithium battery pack representing the service life of the lithium battery is taken as a performance function J (k) of the neural network,
I bt,a (k) Represents the output current of the lithium battery pack at the time of k, and N represents the output current I of the lithium battery pack at the time of k bt,a The total number of sampling values;
by utilizing a BP learning algorithm, when the forward propagation is carried out, an input variable is transmitted to an output layer from an input layer through a hidden layer; when reversely propagating, transmitting the performance function J (k) from the output layer to the input layer through the hidden layer, and adjusting the weight and the threshold of the neural network on line according to a gradient descent method; searching the negative gradient direction of the neural network according to J (k), adding an inertia term which enables the search to be rapidly converged, and optimizing an energy management strategy, wherein a weight value and threshold value adjusting formula is as follows:
where η is the learning rate and α is the inertia coefficient.
7. The real-time energy optimization method for the lithium battery and supercapacitor composite power supply system according to claim 5, wherein the method comprises the following steps: step S7, evaluating the optimization result, and finishing the optimization when the optimization result meets the requirement; and when the fuzzy energy can not meet the requirements, modifying the fuzzy energy management rule, and optimizing according to the steps S1-S6.
8. The real-time energy optimization method for the lithium battery and supercapacitor composite power supply system according to claim 1, wherein the method comprises the following steps: the energy real-time optimization method is completed in ADVISOR software based on MATLAB.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911384077.2A CN111079349B (en) | 2019-12-28 | 2019-12-28 | Energy real-time optimization method for lithium battery and super capacitor composite power supply system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911384077.2A CN111079349B (en) | 2019-12-28 | 2019-12-28 | Energy real-time optimization method for lithium battery and super capacitor composite power supply system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111079349A CN111079349A (en) | 2020-04-28 |
| CN111079349B true CN111079349B (en) | 2023-04-07 |
Family
ID=70319102
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911384077.2A Active CN111079349B (en) | 2019-12-28 | 2019-12-28 | Energy real-time optimization method for lithium battery and super capacitor composite power supply system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111079349B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112434463B (en) * | 2020-10-27 | 2023-04-07 | 中北大学 | Energy management system for vehicle hybrid power supply |
| CN113071507B (en) * | 2021-03-22 | 2022-03-01 | 江铃汽车股份有限公司 | Electric automobile energy management control method based on fuzzy control |
| CN116985646B (en) * | 2023-09-28 | 2024-01-12 | 江西五十铃汽车有限公司 | Vehicle supercapacitor control method, device and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013058568A1 (en) * | 2011-10-18 | 2013-04-25 | 주식회사 제이에스영테크 | Hybrid battery system for an electric vehicle |
| GB201506497D0 (en) * | 2015-04-16 | 2015-06-03 | Oxis Energy Ltd And Cranfield University And Imp Innovations Ltd | Method and apparatus for determining the state of health and state of charge of lithium sulfur batteries |
| CN104682490A (en) * | 2014-12-26 | 2015-06-03 | 智慧城市系统服务（中国）有限公司 | Method and device for controlling battery charging and discharging current |
| CN106347162A (en) * | 2016-11-01 | 2017-01-25 | 大连民族大学 | Hybrid power management and control system and hybrid power management and control method for electric racing car based on super-capacitor |
| CN110231566A (en) * | 2019-06-13 | 2019-09-13 | 华中科技大学鄂州工业技术研究院 | A kind of electric energy managing and control system |
-
2019
- 2019-12-28 CN CN201911384077.2A patent/CN111079349B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013058568A1 (en) * | 2011-10-18 | 2013-04-25 | 주식회사 제이에스영테크 | Hybrid battery system for an electric vehicle |
| CN104682490A (en) * | 2014-12-26 | 2015-06-03 | 智慧城市系统服务（中国）有限公司 | Method and device for controlling battery charging and discharging current |
| GB201506497D0 (en) * | 2015-04-16 | 2015-06-03 | Oxis Energy Ltd And Cranfield University And Imp Innovations Ltd | Method and apparatus for determining the state of health and state of charge of lithium sulfur batteries |
| CN106347162A (en) * | 2016-11-01 | 2017-01-25 | 大连民族大学 | Hybrid power management and control system and hybrid power management and control method for electric racing car based on super-capacitor |
| CN110231566A (en) * | 2019-06-13 | 2019-09-13 | 华中科技大学鄂州工业技术研究院 | A kind of electric energy managing and control system |
Non-Patent Citations (6)
| Title |
|---|
| A.Zenati et al..Impedance measurements combined with the fuzzy logic methodology to assess the SOC and SOH of lithium-ion cells.《VPPC》.2011,全文. * |
| Jonghoon Kim et al..Fuzzy logic-controlled online state-of-health (SOH) prediction in large format LiMn2O4 cell for energy storage system (ESS) applications.《ICIT》.2014,全文. * |
| Ridong Zhang et al..Fuzzy optimal energy management for fuel cell and supercapacitor systems using neural network based driving pattern recognition.《TOFS》.2019,第27卷全文. * |
| Xiao peng et al..A new supercapacitor and Li-on battery system for electric vehicle in ADVISOR.《J.Phys.:Conf.Ser.》.2017,全文. * |
| Xiaohui zhang et al..Fuzzy state machine energy management strategy for hybrid electric UAVs with PV/Fuel cell/battery power system.《Hindawi》.2018,全文. * |
| 肖朋等.复合电源系统电动汽车性能敏感度研究.《华中科技大学学报（自然科学版）》.2018,第46卷(第46期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111079349A (en) | 2020-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110126813B (en) | An energy management method for a vehicle fuel cell hybrid power system | |
| CN111079349B (en) | Energy real-time optimization method for lithium battery and super capacitor composite power supply system | |
| CN111409510B (en) | An energy management method for a hydrogen fuel cell hybrid vehicle | |
| CN109494777B (en) | A kind of energy coordination distribution control method for hybrid energy storage system | |
| CN112287463A (en) | An energy management method for fuel cell vehicles based on deep reinforcement learning algorithm | |
| CN111572369A (en) | Fuel cell hybrid electric vehicle energy management online optimization method based on improved genetic algorithm | |
| CN110303945B (en) | A method for self-adaptive optimal balance control of battery pack power | |
| CN110247451B (en) | Lithium ion power battery pack full-life cycle balance control method | |
| CN117879128B (en) | Energy management system and management strategy of large container ship composite energy system | |
| CN112467717A (en) | Hybrid energy system real-time load distribution method based on fuzzy control | |
| CN114801897A (en) | DDPG algorithm-based fuel cell hybrid power system energy management method | |
| Hu et al. | Energy management strategy for fuel cell hybrid vehicles considering optimal performance of multiple energy sources combined with ant colony algorithm | |
| CN118195828A (en) | A composite energy storage method for optimizing energy management of electric vehicles | |
| CN117644783A (en) | Fuel cell automobile energy management method combining working condition prediction and reinforcement learning | |
| CN115983134B (en) | Battery power state prediction method and system based on neural network | |
| CN114347866B (en) | A fuel cell energy management method based on VMD processing | |
| CN112946480A (en) | Lithium battery circuit model simplification method for improving SOC estimation real-time performance | |
| CN115848217B (en) | An energy management method based on multi-energy modules | |
| Vonteddu et al. | Dynamic Modeling for Hybrid Electric Vehicles with its Performance Optimization Using Different Control Strategies and Energy Management Systems | |
| Feng et al. | Research on On-line identification of equivalent circuit model parameters and soc estimation for lithium-ion batteries | |
| Lei et al. | Energy management strategy based on adaptive wavelet analysis for HESS | |
| Bin et al. | Research on SOC Estimation and Energy Cooperative Control for Electric Vehicles | |
| CN114179781B (en) | Plug-in hybrid electric vehicle real-time control optimization method and system | |
| CN115610246B (en) | Topological structure and energy distribution method of extended-range hydrogen fuel cell hybrid electric vehicle | |
| Xiao et al. | A collaborative energy management strategy based on multi-agent reinforcement learning for fuel cell hybrid electric vehicles |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240906 Address after: 312000 Zhejiang Province, Shaoxing City, Shangyu District, Cao'e Street, No. 2288 Jiangxi Road, Zhejiang University Network New Technology Park, A1 Building, 5th Floor, 513 (Address Declaration) Patentee after: Zhejiang Weiranyun New Energy Technology Co.,Ltd. Country or region after: China Address before: 312000 Zhejiang University network new science and Technology Park, No. 2288, Shangyu District, Shaoxing City, Zhejiang Province Patentee before: Shaoxing Shangyu District Institute of Technology Country or region before: China |