CN107301266A - A kind of ferric phosphate lithium cell LOC evaluation methods and system - Google Patents
A kind of ferric phosphate lithium cell LOC evaluation methods and system Download PDFInfo
- Publication number
- CN107301266A CN107301266A CN201710339846.1A CN201710339846A CN107301266A CN 107301266 A CN107301266 A CN 107301266A CN 201710339846 A CN201710339846 A CN 201710339846A CN 107301266 A CN107301266 A CN 107301266A
- Authority
- CN
- China
- Prior art keywords
- lithium cell
- ferric phosphate
- network
- phosphate lithium
- wavelet
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Secondary Cells (AREA)
Abstract
A kind of ferric phosphate lithium cell LOC evaluation methods, comprise the following steps:Set up wavelet-neural network model;Build wavelet neural network;Obtain the input parameter relevant with ferric phosphate lithium cell LOC;Input parameter is inputted into wavelet neural network and carries out data processing, so as to export ferric phosphate lithium cell LOC values.The invention also discloses a kind of ferric phosphate lithium cell LOC estimating systems corresponding with the evaluation method.The present invention can with it is accurate and effective, reliably ferric phosphate lithium cell LOC is estimated, it can be applied to batteries of electric automobile group management of charging and discharging and associated batteries industry, and easily realize, have a good application prospect and market value.
Description
Technical field
The present invention relates to battery LOC (pot life) evaluation methods field, more particularly, to a kind of LiFePO4 electricity
Pond LOC evaluation methods, it can be applied to batteries of electric automobile group management of charging and discharging and associated batteries industry.
Background technology
Electric automobile as future transportation instrument development trend, with environmental protection, energy-conservation, it is light the features such as by people green grass or young crops
Look at, as the battery of key technology part, its performance directly influences the quality of vehicle performance.Battery is the dynamic of electric automobile
Power source, its life-span and price relation are to the cost of electric automobile, and its stored energy capacitance and power density decide electric automobile
Distance travelled and speed, and can these factors directly affect electric automobile really move towards market, therefore, exploitation long lifespan,
High-performance, the battery of low cost are the only way which must be passed of Development of EV.
Electrokinetic cell, as stored energy power source, is the core component of electric automobile, current super capacitor and lithium ion battery
It is the main stored energy power source applied on electric automobile.Super capacitor is influenceed by its energy density and price, in pure electric vehicle
On application be limited by very large.Cell positive material for LiFePO4 lithium battery compared to other batteries, it has
Specific energy height, voltage height, low self-discharge rate, discharge and recharge long lifespan, memory-less effect, operating temperature range are wide, pollution-free and pacify
Complete reliable the advantages of.
As a major issue of control and the management of battery system, the key for being also electric car traveling anxiety disorder is exactly
How long battery can use if there remains.The present invention proposes LOC pot life parameters, by with critically important realistic meaning and
Economic value.
In the actual moving process of electric automobile, the operating mode residing for battery is extremely complex, can survey parameter voltages, electric current,
The relation between LOC such as temperature is all that complicated non-linear relation, particularly electric automobile are starting, accelerated or climbing rank
Section, battery be heavy-current discharge state, voltage, electric current change it is very violent, fluctuate it is larger, while varying environment temperature is to electricity
The performance in pond also has a great impact,
The present invention proposes to use scalable dimension Wavelet Neural Network by analyzing influence battery LOC each side factor
Network method solves the forecasting problem of ferric phosphate lithium cell pot life.
The content of the invention
In order to solve the above-mentioned technical problem, realize it is accurate and effective, reliably ferric phosphate lithium cell LOC is estimated, this
Invention provides a kind of ferric phosphate lithium cell LOC evaluation methods and system, and the technical scheme that its problem of present invention solution is used is:
The invention provides a kind of ferric phosphate lithium cell LOC evaluation methods, the evaluation method comprises the following steps:
S1:Set up wavelet-neural network model;
S2:Build wavelet neural network;
S3:Obtain the input parameter relevant with ferric phosphate lithium cell LOC;
S4:Input parameter is inputted into wavelet neural network and carries out data processing, so as to export ferric phosphate lithium cell LOC
Value.
Further, the topological structure of wavelet-neural network model includes input layer, hidden layer and output layer.
Further, the node-node transmission function of hidden layer is wavelet basis function.
Further, wavelet basis function uses Morlet functions, and its mathematical formulae is:
Further, the step S2 includes:
(1) initial network is set up, wherein the initial network is made up of yardstick for j wavelet frame;
(2) be trained, and whether error in judgement becomes big in the training process, record error become it is big before weights simultaneously
As the final weights of current network, meanwhile, judge whether precision meets requirement, the deconditioning if precision meets requirement, if
Precision is unsatisfactory for requiring, is incorporated to yardstick and is j+1 sub-network, and makes the desired output of this grade of network be the mistake of upper level network
Difference, the i.e. error to upper level network learn;
(3) by that analogy, when being incorporated to m-th of sub-network, precision, which is met, terminates training when requiring, now instructed
The weights for the whole network perfected, wavelet neural network, which is built, to be completed.
Further, input parameter includes electric current, voltage, temperature, discharge rate and cycle-index.
In addition, present invention also offers a kind of system for realizing ferric phosphate lithium cell LOC evaluation methods, the system bag
Include:
Modeling unit, for setting up wavelet-neural network model;
Network struction unit, for building wavelet neural network;
Acquiring unit, for obtaining the input parameter relevant with ferric phosphate lithium cell LOC;
Processing unit, data processing is carried out for the input parameter to be inputted into wavelet neural network, so as to export
Ferric phosphate lithium cell LOC values.
Further, the topological structure of wavelet-neural network model includes input layer, hidden layer and output layer.
Further, the node-node transmission function of hidden layer is wavelet basis function, and the wavelet basis function uses Morlet letters
Count, its mathematical formulae is:
Further, input parameter includes electric current, voltage, temperature, discharge rate and cycle-index.
Beneficial effects of the present invention are:
1st, the comprehensive selection of scale parameter and translation parameters.Generally, the multidimensional wavelet network based on framework is to every
One-dimensional input signal goes to approach to the Nonlinear Mapping between output signal using different wavelet frames, and this does not only result in each frame
Larger redundancy is there may be between frame, and with the growth of input dimension, the number of hidden nodes of network will exponentially increase
It is long.The present invention to each dimension input signal by carrying out integrating time-domain analysis, and in the case where scale parameter is determined, selection is suitable
So that small echo stretches, system is that Tight wavelet frames under single scale can cover each dimensional input vector to displacement parameter, so as to use one
Individual Tight wavelet frames go to approach each dimension input signal to the Nonlinear Mapping between output signal.So, difference is not only reduced
Redundancy between wavelet frame, and network node in hidden layer also will not with input dimension increase and exponentially increase
It is long.
2nd, the adaptive of hidden layer node should determine that.The present invention construct based on the tight branch radial direction wavelet frame of single scale multidimensional
Adaptive single order self feed back wavelet neural network includes initial network and adaptive according to the requirement of precision in the training process
The sub-network being incorporated to.Initial network and each sub-network are the feedforward neural network of single hidden layer, and the hidden layer of networks at different levels is equal
The tight branch radial direction wavelet frame of multidimensional under single scale is constituted, and yardstick increases step by step, per primary network station to primary network station thereon
Error learnt, and the training of the sub-network parameter on being newly incorporated to does not influence the parameter of the network trained, until
Untill meeting required precision.The node in hidden layer of network can be thus adaptively determined according to the requirement of precision, is made hidden
The determination of the number containing node layer is evidence-based, largely optimizes the structure of network.
3rd, the application of single order self feed back structure.The hidden layer node of network uses single order self feed back structure, by introducing one
Rank reflexive feedback link reflects the dynamic characteristic of system, thus can be without leading to as traditional multiplayer dynamic forward networks
Cross the input to network and output makees m ranks and n ranks time delay and feeds back to input to reflect system by the output of network respectively
Dynamic characteristic, so just largely reduces the input number of nodes of network.Because the input of hidden layer node is by working as
The input at preceding moment and the output composition of previous moment, and previous moment is output as the function of moment input, previous moment
The input output comprising previous moment again again, unlimited recursion is thusly-formed, so its memory to information is unlimited, so
Just the dynamic characteristic of system can preferably be reflected on the premise of network inputs node is not increased.
Brief description of the drawings
Fig. 1 is a kind of step FB(flow block) of ferric phosphate lithium cell LOC evaluation methods of the present invention;
Fig. 2 is the topological structure of preferred embodiment of the present invention wavelet-neural network model;
Fig. 3 is the flow chart that the preferred embodiment of the present invention builds wavelet neural network;
Fig. 4 is an a kind of specific embodiment structural frames schematic diagram of ferric phosphate lithium cell LOC estimating systems of the invention.
Embodiment
The present invention is described in detail with embodiment with reference to the accompanying drawings, and wherein specific embodiment should not be construed as to this
The limitation of invention scope.
The LOC of battery is the direct basis of the sustainable traveling of electric automobile, and therefore, it is accurate that electric automobile is estimated it
Property, reliability propose higher requirement.First, we are necessary to learn about ferric phosphate lithium cell LOC theoretical formula:
LOC=Qn×(SOC-DZSOC)/V.The state-of-charge SOC of life-span LOC and battery, dead band capacity in cell single cycle
DZSOC, battery rated capacity Qn, discharge rate V it is relevant, while considering that various operating modes and stress should be using weightings in realistic model
Coefficient is modified processing to above formula.
In order to it is more accurate and effective, reliably ferric phosphate lithium cell LOC is estimated, the present invention proposes a kind of phosphoric acid
Lithium iron battery LOC evaluation methods, steps flow chart block diagram shown in reference picture 1, in one embodiment, the evaluation method include with
Lower step:
S1:Set up wavelet-neural network model;
S2:Build wavelet neural network;
S3:Obtain the input parameter relevant with ferric phosphate lithium cell LOC;
S4:The input parameter is inputted into wavelet neural network and carries out data processing, so as to export LiFePO4 electricity
Pond LOC values.
Further, the wavelet-neural network model set up in step 1 is to be based on BP neural network model, by BP nerves
The node-node transmission function of hidden layer in network model is replaced with wavelet basis function, and the BP neural network model after replacement is with regard to structure
Into wavelet-neural network model.Advantage of this is that, wavelet transformation is to stretch to carry out signal by translation and yardstick
Multiscale analysis, it so can more effectively extract the local message of signal;For neutral net, itself have good
Adaptivity, preferable ability of self-teaching and the features such as stronger fault-tolerance, can use it to approach most of function.Should
What forecast model was preferred to use in the present invention is three layers of pre- geodesic structures of WNN, i.e., including input layer, hidden layer and output layer.Cause
Can not Accurate Prediction for the forecast model less than three-decker;The forecast model that structure is more than three layers can correspondingly increase calculating
Amount.Therefore, in a preferred embodiment, the topological structure of wavelet-neural network model is as shown in Fig. 2 wherein, wavelet basis function
It is preferred to use Morlet functions, its mathematical formulae is:
Reference picture 2, the mathematical modeling for the wavelet neural network set up can be expressed as:
Wherein, xi(t) it is i-th of input of network, Ij(t) it is j-th of wavelet basis function input sum, Sj(t) it is jth
The output of individual small echo.ψ is wavelet function,And yk(t) it is k-th of output node of network, ek(t) it is model error, it is defeated
The weights for entering layer to hidden layer areThe weights of hidden layer to output layer are For 1, hidden layer node, which passes through, introduces one
The reflexive feedback link of rank reflects the dynamic characteristic of system, thus can be without traditional multilayer dynamic forward net like that by right
The input and output of network make m ranks and n ranks time delay and feed back to input to reflect the dynamic of system by the output of network respectively
Characteristic, largely reduces the input number of nodes of network.The input of hidden layer node is by the input at current time with before
The output composition at one moment, and previous moment is output as the function of moment input, the input of previous moment is included again again before
The output at one moment, is thusly-formed unlimited recursion, so its memory to information is unlimited.It can be seen that, with multilayer dynamic forward
Net is compared, and the dynamic characteristic of system can be preferably reflected using first-order lag structure, and does not increase the input number of nodes of network.
Further, in a preferred embodiment, the flow of wavelet neural network is built in step 2 as shown in figure 3, specific
It may include:
(1) set up initial network, wherein the initial network is made up of yardstick for j wavelet frame, wherein j be one compared with
Small integer.When j takes 1 or 2, training precision is generally very low, therefore the yardstick of initial network can be taken as 3,4 or bigger
Integer;
(2) it is trained, and whether error in judgement becomes big in the training process, error is gradually reduced directly during training
To keeping substantially constant, when error becomes it is big when then show network training excessively, if being further continued for training, can make the error of network after
Continuous increase, now record error become it is big before weights and as the final weights of current network, meanwhile, whether judge precision
Meet and require, the deconditioning if precision meets requirement is incorporated to the sub-network that yardstick is j+1 if precision is unsatisfactory for requiring, and
The desired output for making this grade of network is the error of upper level network, i.e. the error to upper level network learns;
(3) by that analogy, when being incorporated to m-th of sub-network, precision, which is met, terminates training when requiring, now instructed
The weights for the whole network perfected, wavelet neural network, which is built, to be completed.
The adaptive single order self feed back wavelet neural based on the tight branch radial direction wavelet frame of single scale multidimensional that the present invention is constructed
Network includes initial network and the sub-network being adaptively incorporated to according to the requirement of precision in the training process.Initial network and each
Sub-network is the feedforward neural network of single hidden layer, and the hidden layer of networks at different levels is radially small by the tight branch of multidimensional under single scale
Ripple framework is constituted, and yardstick increases step by step, and the error of primary network station thereon is learnt per primary network station, and to being newly incorporated to
The training of sub-network parameter do not influence the parameter of the network trained, untill meeting required precision.Thus can be with
The node in hidden layer of network is adaptively determined according to the requirement of precision, makes the determination of node in hidden layer evidence-based,
The structure of network is optimized in very big degree.
Further, in step 3, obtaining the input parameter relevant with ferric phosphate lithium cell LOC may include:Electric current, electricity
Pressure, temperature, discharge rate and cycle-index.
Further, in step 4, the input parameter relevant with ferric phosphate lithium cell LOC step 3 obtained input to
Built through step 2 in the wavelet neural network completed and carry out data processing, so as to export ferric phosphate lithium cell LOC values.
In addition, present invention also offers a kind of ferric phosphate lithium cell LOC estimating systems, as shown in figure 4, being the one of the system
Specific embodiment structural frames schematic diagram, the system includes:
Modeling unit, for setting up wavelet-neural network model;
Network struction unit, for building wavelet neural network;
Acquiring unit, for obtaining the input parameter relevant with ferric phosphate lithium cell LOC;
Processing unit, data processing is carried out for input parameter to be inputted into wavelet neural network, so as to export phosphoric acid
Lithium iron battery LOC values.
Further, modeling unit set up wavelet-neural network model topological structure include input layer, hidden layer and
Output layer.
Further, the node-node transmission function of the hidden layer is wavelet basis function, and the wavelet basis function uses Morlet
Function, its mathematical formulae is:
Further, the flow of network struction cell formation wavelet neural network is as described in step 2.
Further, acquiring unit obtains the input parameter relevant with ferric phosphate lithium cell LOC, it may include:Electric current, electricity
Pressure, temperature, discharge rate and cycle-index.
Further, the input parameter input relevant with ferric phosphate lithium cell LOC that processing unit obtains acquiring unit
Data processing is carried out into wavelet neural network, so as to export ferric phosphate lithium cell LOC values.
It is described above, simply presently preferred embodiments of the present invention, the invention is not limited in above-mentioned embodiment, as long as
It reaches the technique effect of the present invention with identical means, should all belong to protection scope of the present invention.In the protection model of the present invention
Its technical scheme and/or embodiment can have a variety of modifications and variations in enclosing.
Claims (10)
1. a kind of ferric phosphate lithium cell LOC evaluation methods, it is characterised in that comprise the following steps:
S1:Set up wavelet-neural network model;
S2:Build wavelet neural network;
S3:Obtain the input parameter relevant with ferric phosphate lithium cell LOC;
S4:The input parameter is inputted into wavelet neural network and carries out data processing, so as to export ferric phosphate lithium cell LOC
Value.
2. a kind of ferric phosphate lithium cell LOC evaluation methods according to claim 1, it is characterised in that the wavelet neural
The topological structure of network model includes input layer, hidden layer and output layer.
3. a kind of ferric phosphate lithium cell LOC evaluation methods according to claim 2, it is characterised in that the hidden layer
Node-node transmission function is wavelet basis function.
4. a kind of ferric phosphate lithium cell LOC evaluation methods according to claim 3, it is characterised in that the wavelet basis letter
Number uses Morlet functions, and its mathematical formulae is:
5. a kind of ferric phosphate lithium cell LOC evaluation methods according to claim 1, it is characterised in that the step S2 bags
Include:
(1) initial network is set up, wherein the initial network is made up of yardstick for j wavelet frame;
(2) it is trained, and whether error in judgement becomes big in the training process, records the weights before error becomes greatly and conduct
The final weights of current network, meanwhile, judge whether precision meets requirement, the deconditioning if precision meets requirement, if precision
It is unsatisfactory for requiring, is incorporated to yardstick and is j+1 sub-network, and make the desired output of this grade of network be the error of upper level network, i.e.,
Error to upper level network learns;
(3) by that analogy, when being incorporated to m-th of sub-network, precision, which is met, terminates training when requiring, now trained
Whole network weights, wavelet neural network build complete.
6. a kind of ferric phosphate lithium cell LOC evaluation methods according to claim 1, it is characterised in that the input parameter
Including electric current, voltage, temperature, discharge rate and cycle-index.
7. a kind of ferric phosphate lithium cell LOC estimating systems, it is characterised in that the system includes:
Modeling unit, for setting up wavelet-neural network model;
Network struction unit, for building wavelet neural network;
Acquiring unit, for obtaining the input parameter relevant with ferric phosphate lithium cell LOC;
Processing unit, data processing is carried out for the input parameter to be inputted into wavelet neural network, so as to export phosphoric acid
Lithium iron battery LOC values.
8. a kind of ferric phosphate lithium cell LOC estimating systems according to claim 7, it is characterised in that the wavelet neural
The topological structure of network model includes input layer, hidden layer and output layer.
9. a kind of ferric phosphate lithium cell LOC estimating systems according to claim 7, it is characterised in that the node of the hidden layer
Transfer function is wavelet basis function, and the wavelet basis function uses Morlet functions, and its mathematical formulae is:
10. a kind of system for realizing ferric phosphate lithium cell LOC evaluation methods according to claim 7, its feature exists
In the input parameter includes electric current, voltage, temperature, discharge rate and cycle-index.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710339846.1A CN107301266B (en) | 2017-05-15 | 2017-05-15 | LOC estimation method and system for lithium iron phosphate battery |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710339846.1A CN107301266B (en) | 2017-05-15 | 2017-05-15 | LOC estimation method and system for lithium iron phosphate battery |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107301266A true CN107301266A (en) | 2017-10-27 |
CN107301266B CN107301266B (en) | 2020-11-10 |
Family
ID=60137219
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710339846.1A Expired - Fee Related CN107301266B (en) | 2017-05-15 | 2017-05-15 | LOC estimation method and system for lithium iron phosphate battery |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107301266B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110007235A (en) * | 2019-03-24 | 2019-07-12 | 天津大学青岛海洋技术研究院 | A kind of accumulator of electric car SOC on-line prediction method |
CN112818594A (en) * | 2021-01-28 | 2021-05-18 | 温州大学 | Multi-objective optimization battery pack structure method based on neural network |
CN113049961A (en) * | 2021-02-26 | 2021-06-29 | 佛山职业技术学院 | DZSOC algorithm of lithium iron phosphate battery |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103716177A (en) * | 2013-11-18 | 2014-04-09 | 国家电网公司 | Security risk assessment method and apparatus |
US8760115B2 (en) * | 2009-08-20 | 2014-06-24 | GM Global Technology Operations LLC | Method for charging a plug-in electric vehicle |
US20140244225A1 (en) * | 2013-02-24 | 2014-08-28 | The University Of Connecticut | Battery state of charge tracking, equivalent circuit selection and benchmarking |
CN106443453A (en) * | 2016-07-04 | 2017-02-22 | 陈逸涵 | Lithium battery SOC estimation method based on BP neural network |
CN106650060A (en) * | 2016-12-08 | 2017-05-10 | 国网青海省电力公司 | Prediction method of internal resistance attenuation coefficient for photovoltaic cells |
-
2017
- 2017-05-15 CN CN201710339846.1A patent/CN107301266B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8760115B2 (en) * | 2009-08-20 | 2014-06-24 | GM Global Technology Operations LLC | Method for charging a plug-in electric vehicle |
US20140244225A1 (en) * | 2013-02-24 | 2014-08-28 | The University Of Connecticut | Battery state of charge tracking, equivalent circuit selection and benchmarking |
CN103716177A (en) * | 2013-11-18 | 2014-04-09 | 国家电网公司 | Security risk assessment method and apparatus |
CN106443453A (en) * | 2016-07-04 | 2017-02-22 | 陈逸涵 | Lithium battery SOC estimation method based on BP neural network |
CN106650060A (en) * | 2016-12-08 | 2017-05-10 | 国网青海省电力公司 | Prediction method of internal resistance attenuation coefficient for photovoltaic cells |
Non-Patent Citations (3)
Title |
---|
刘勇智 等: "基于小波神经网络的航空蓄电池容量预测", 《电源技术研究与设计》 * |
刘学鹏 等: "磷酸铁锂电池组三级管理系统的开发与应用", 《电源技术研究与设计》 * |
章琴 等: "基于改进的Morlet小波变换的雷达信号特征提取", 《微型机与应用》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110007235A (en) * | 2019-03-24 | 2019-07-12 | 天津大学青岛海洋技术研究院 | A kind of accumulator of electric car SOC on-line prediction method |
CN112818594A (en) * | 2021-01-28 | 2021-05-18 | 温州大学 | Multi-objective optimization battery pack structure method based on neural network |
CN112818594B (en) * | 2021-01-28 | 2023-05-30 | 温州大学 | Multi-objective battery pack structure optimizing method based on neural network |
CN113049961A (en) * | 2021-02-26 | 2021-06-29 | 佛山职业技术学院 | DZSOC algorithm of lithium iron phosphate battery |
CN113049961B (en) * | 2021-02-26 | 2022-07-19 | 佛山职业技术学院 | DZSOC algorithm of lithium iron phosphate battery |
Also Published As
Publication number | Publication date |
---|---|
CN107301266B (en) | 2020-11-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xiong et al. | Towards a smarter hybrid energy storage system based on battery and ultracapacitor-A critical review on topology and energy management | |
CN103954915B (en) | Lithium ion battery remaining life indirect prediction method based on probability integration | |
CN108519556A (en) | A kind of lithium ion battery SOC prediction techniques based on Recognition with Recurrent Neural Network | |
CN112331941B (en) | Cloud auxiliary battery management system and method | |
Xu et al. | A soft actor-critic-based energy management strategy for electric vehicles with hybrid energy storage systems | |
CN104617623B (en) | A kind of electric automobile power battery group balance control method | |
CN108226809A (en) | A kind of multi-model and battery SOC evaluation method | |
Wu et al. | The SOC estimation of power Li-Ion battery based on ANFIS model | |
CN104217258B (en) | A kind of electric load sigma-t Forecasting Methodology | |
Capizzi et al. | Recurrent neural network-based control strategy for battery energy storage in generation systems with intermittent renewable energy sources | |
CN108537337A (en) | Lithium ion battery SOC prediction techniques based on optimization depth belief network | |
CN110007235A (en) | A kind of accumulator of electric car SOC on-line prediction method | |
CN110941929A (en) | Battery health state assessment method based on ARMA and Elman neural network combined modeling | |
CN102569922A (en) | Improved storage battery SOC estimation method based on consistency of unit cell | |
Zhang et al. | Intelligent state of charge estimation of battery pack based on particle swarm optimization algorithm improved radical basis function neural network | |
Zhang et al. | State‐of‐charge estimation of lithium‐ion battery pack based on improved RBF neural networks | |
CN110247451A (en) | A kind of lithium-ion-power cell group life cycle management balance control method | |
CN107301266A (en) | A kind of ferric phosphate lithium cell LOC evaluation methods and system | |
CN103983919A (en) | Battery life prediction method based on GM (1, N) gray model | |
CN108009585A (en) | Lead-acid battery health status Forecasting Methodology based on local message fusion | |
CN109214565A (en) | A kind of subregion system loading prediction technique suitable for the scheduling of bulk power grid subregion | |
CN113312807B (en) | Electrolyte formula recommendation method based on lithium battery performance simulation environment | |
CN110297452A (en) | A kind of equal balance system of the adjacent type of battery group and its forecast Control Algorithm | |
CN104915788A (en) | Method for electric power system dynamic economic dispatching with correlation of multiple wind power plants being considered | |
CN113344283A (en) | Energy internet new energy consumption capacity assessment method based on edge intelligence |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201110 |