CN104881001A - Energy storage battery management system based on deep learning network - Google Patents

Energy storage battery management system based on deep learning network Download PDF

Info

Publication number
CN104881001A
CN104881001A CN201410169036.2A CN201410169036A CN104881001A CN 104881001 A CN104881001 A CN 104881001A CN 201410169036 A CN201410169036 A CN 201410169036A CN 104881001 A CN104881001 A CN 104881001A
Authority
CN
China
Prior art keywords
management unit
module
battery management
battery
neural network
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.)
Withdrawn
Application number
CN201410169036.2A
Other languages
Chinese (zh)
Inventor
陈宗海
汪玉洁
张陈斌
武骥
桂旺胜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ANHUI GVB RENEWABLE ENERGY TECHNOLOGY Co Ltd
Original Assignee
ANHUI GVB RENEWABLE ENERGY TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by ANHUI GVB RENEWABLE ENERGY TECHNOLOGY Co Ltd filed Critical ANHUI GVB RENEWABLE ENERGY TECHNOLOGY Co Ltd
Priority to CN201410169036.2A priority Critical patent/CN104881001A/en
Publication of CN104881001A publication Critical patent/CN104881001A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/418Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/08Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers from or to individual record carriers, e.g. punched card, memory card, integrated circuit [IC] card or smart card
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

The invention discloses an energy storage battery management system based on a deep learning network. The energy storage battery management system comprises a battery management unit BMU, a central management unit CMU, an upper computer, a 4V storage battery, a Hall sensor and a DC load, wherein the 24V storage battery is electrically connected with the battery management unit BMU and the central management unit CMU, the central management unit CMU communicates with the battery management unit BMU through a CAN bus, and the upper computer is in communication connection with the central management unit CMU through an RS485. Compared with the prior art, the energy storage battery management system based on the deep learning network is provided for realizing on-line monitoring of an energy storage battery pack, and improves estimation accuracy of battery management system state-of-charge SOC and storage battery SOH through a deep neural network model and a learning algorithm thereof.

Description

A kind of energy storage battery management system based on degree of depth learning network
Technical field
The present invention relates to energy management field, be specially a kind of energy storage battery management system based on degree of depth learning network.
Background technology
Along with the develop rapidly of Power Electronic Technique and computer technology, distributed power generation, as one of method solving the problems such as energy economy & environment that conventional power generation systems brings, is more and more subject to people's attention.But, the active type distribution that distributed power source and user mix and formed is that electric system brings new challenge, for solving the problem of distributed power source access, coordinate the contradiction of bulk power grid and distributed power source, abundant excavation distributed power generation is the value brought of electrical network and user and benefit, and micro-capacitance sensor technology is arisen at the historic moment.
Electricity storage technology weighs for the basic function realizing micro-capacitance sensor very much, and high performance accumulator system can ensure power supply quality and the continued reliability of electric power system, improves the comprehensive utilization ratio of electric energy.Therefore, perfect energy storage battery management system is the powerful guarantee that micro-capacitance sensor realizes himself function.Lithium ion battery due to its have specific energy large, have extended cycle life, self-discharge rate is little, memory-less effect and the feature such as pollution-free, has become the study hotspot of countries nowadays energy storage field.But, the high energy ratio performance of lithium ion battery is that it brings the hidden danger of secure context, and incorrect use-pattern will have a strong impact on the use safety of battery, therefore must strictly according to battery parameter management battery, prevent it from overcharging, cross and put and excess temperature, thus ensure its security performance.Because the charge and discharge platform of lithium battery is very smooth, simultaneously, the charge and discharge process of lithium battery is an extremely complicated process, charge-discharge magnification and the factor such as environment temperature, humidity all can have influence on decay and the serviceable life of pack total capacities, use conventional method to carry out state estimation to lithium battery to be difficult to obtain estimated result accurately, this patent provides a kind of method based on degree of depth network for improving the accuracy of battery management system state estimation.
Summary of the invention
The invention provides a kind of energy storage battery management system based on degree of depth learning network, for the monitoring and control of micro-capacitance sensor energy-storage battery group particularly ferric phosphate lithium cell.
For achieving the above object, the invention provides following technical scheme:
Based on an energy storage battery management system for degree of depth learning network, comprising: battery management unit BMU, central management unit CMU, host computer, 24V accumulator, Hall element and DC load; Wherein 24V accumulator is electrically connected with battery management unit BMU and central management unit CMU, carries out communication between central management unit CMU and battery management unit BMU by CAN, is connected between host computer and central management unit CMU by RS485 communication.
Described battery management unit BMU comprises: malfunction coefficient module, data acquisition module, balance module, CAN and power module, data acquisition module carries out voltage, temperature acquisition send information to central management unit CMU by CAN, carry out battery cell equilibrium and failure code display, preferably, described malfunction coefficient module is made up of seven segmentation charactrons; Described data acquisition module is made up of voltage monitoring chip LTC6802, NTC thermistor and buffer circuit; Described balance module is made up of matrix switch, flyback DC/DC transducer and balance controller; Described CAN receives and dispatches TJA1050 by high-speed CAN, two-channel digital isolating chip IS07221A is formed; Described power module is made up of voltage stabilizing PWB2405CS and IB00505S.
Described central management unit CMU comprises: data memory module, current detection module, RS485 communication module, CAN and power module, and preferably, described data memory module is made up of Micro SD card and level transferring chip ADG3308; Current detection module is made up of Hall current sensor, operational amplifier; Described RS485 communication module is made up of differential data transponder chip SN65LBC184D; Described CAN is by high-speed CAN transceiver TJA1050, and two-channel digital isolating chip IS07221A is formed; Described power module is made up of voltage stabilizing chip PWA2412MD, K7805, REG1117 and IB00505S.
Described host computer comprises: HMI touch-screen.
The state-of-charge SOC of battery of the present invention and health status (accumulator SOH) estimation are realized by deep neural network and learning algorithm thereof; Described deep neural network construction step is:
1, input layer and output layer neuron is determined: using battery total capacity, voltage, electric current, temperature and the SOC in k-1 moment as the input layer of neural network; Using the output layer neuron of SOC and SOH of battery as neural network.
2, determine activation function, the neural network hidden layer number of plies and hidden layer neuron number, thus set up neural network.
3, adjust network parameter, train the neural network created, estimate SOC and SOH.
Compared with prior art, the invention provides a kind of energy storage battery management system based on degree of depth learning network, for realizing the on-line monitoring to energy-storage battery group, by deep neural network model and learning algorithm thereof, improve the accuracy that battery management system state-of-charge SOC and accumulator SOH estimates.
Accompanying drawing explanation
Fig. 1 is that the present invention realizes the state-of-charge SOC of battery and the deep neural network structural drawing of health status (accumulator SOH) estimating algorithm.
A kind of energy storage battery management system block diagram based on degree of depth learning network that Fig. 2 provides for the embodiment of the present invention.
Fig. 3 is the structured flowchart of a kind of embodiment of a kind of energy storage battery management system based on degree of depth learning network.
Embodiment
The technological means realized to make the present invention, creation characteristic, reaching object and effect is easy to understand, below in conjunction with concrete diagram, setting forth the present invention further.
A kind of energy storage battery management system schematic diagram based on degree of depth learning network, as shown in Figure 2, comprise: battery management unit BMU, central management unit CMU, host computer, 24V accumulator, Hall element and DC load, in a kind of embodiment of a kind of energy storage battery management system based on degree of depth learning network of the present invention, as shown in Figure 3, described DC load is consumer; Wherein 24V accumulator is electrically connected with battery management unit BMU and central management unit CMU, for powering for battery management unit BMU and central management unit CMU, central management unit CMU carries out current detecting by the Hall element be serially connected between energy-storage battery and DC load, carry out communication by CAN between central management unit CMU and battery management unit BMU, between host computer and central management unit CMU, carry out communication by RS485.
Described battery management unit BMU comprises: malfunction coefficient module, data acquisition module, balance module, CAN and power module, data acquisition module carries out voltage, temperature acquisition send information to central management unit CMU by CAN, carries out the balanced and failure code of battery cell and shows.Further, described malfunction coefficient module is made up of seven segmentation charactrons, for completing the display of failure code; Described data acquisition module is made up of voltage monitoring chip LTC6802, NTC thermistor and buffer circuit, for completing monomer voltage and temperature acquisition; Described balance module is made up of, for realizing battery pack balancing matrix switch, flyback DC/DC transducer and balance controller; Described CAN is made up of, for realizing CAN communication high-speed CAN transceiver TJA1050, two-channel digital isolating chip IS07221A; Described power module is made up of voltage stabilizing chip PWB2405CS and IB00505S, for other module for power supply for battery management unit BMU.
Described central management unit CMU comprises: data memory module, current detection module, RS485 communication module, CAN and power module, and function is: carry out current detecting, electric battery carried out to state estimation, carry out data storage and undertaken alternately by RS485 and host computer; Further, described data memory module is made up of Micro SD card and level transferring chip ADG3308, stores for the critical data produced in operational process; Current detection module is made up of Hall current sensor, operational amplifier, for realizing current acquisition; Described RS485 communication module is made up of differential data transponder chip SN65LBC184D, for realizing the communication of central management unit CMU and HMI touch-screen; Described CAN is by high-speed CAN transceiver TJA1050, and two-channel digital isolating chip IS07221A is formed, for realizing CAN communication; Described power module is made up of voltage stabilizing chip PWA2412MD, K7805, REG1117 and IB00505S, for other module for power supply for central management unit CMU.
Described host computer comprises: HMI touch-screen, and function realizes battery parameter configuration, the status information of Real time displaying battery.
After system electrification, first battery management unit BMU carries out self-inspection, if system discovery fault, then failure message is presented at malfunction coefficient module; If system is normal, then battery cell voltage and temperature is gathered, and monitor CAN by CAN, after finding that there is the CAN request message from central management unit CMU, start to send battery cell infomational message to central management unit CMU.
After central management unit CMU powers on, first need to utilize host computer to carry out parameter configuration to system, after having configured, central management unit CMU can send request message by CAN circulation to battery management unit BMU, receive the battery cell infomational message from battery management unit BMU, central management unit CMU is by Hall element measure loop electric current and utilize deep neural network algorithm to carry out battery status estimation simultaneously.Central management unit CMU also can carry out overcharging in the process run, cross put, the fault diagnosis such as excess temperature the status information of battery and failure message are fed back to user by human-computer interaction interface, on the other hand, the critical data of battery is stored in Micro SD card by central management unit CMU, so that maintainer carries out data analysis.
As shown in Figure 1, the state-of-charge SOC of battery of the present invention and health status (accumulator SOH) estimation are realized by deep neural network and learning algorithm thereof; Described deep neural network construction step is:
1, input layer and output layer neuron is determined: using battery total capacity, voltage, electric current, temperature and the SOC in k-1 moment as the input layer of neural network; Using the output layer neuron of SOC and SOH of battery as neural network.
2, determine activation function, the neural network hidden layer number of plies and hidden layer neuron number, thus set up neural network.
3, adjust network parameter, train the neural network created, estimate SOC and SOH.
To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and when not deviating from spirit of the present invention or essential characteristic, the present invention can be realized in other specific forms.Therefore, no matter from which point, all should embodiment be regarded as exemplary, and be nonrestrictive, scope of the present invention is limited by claims instead of above-mentioned explanation, and all changes be therefore intended in the implication of the equivalency by dropping on claim and scope are included in the present invention.Any Reference numeral in claim should be considered as the claim involved by limiting.

Claims (2)

1. based on an energy storage battery management system for degree of depth learning network, it is characterized in that, comprising: battery management unit BMU, central management unit CMU, host computer, 24V accumulator, Hall element and DC load; Wherein 24V accumulator is electrically connected with battery management unit BMU and central management unit CMU, carries out communication between central management unit CMU and battery management unit BMU by CAN, is connected between host computer and central management unit CMU by RS485 communication; Described battery management unit BMU comprises: malfunction coefficient module, data acquisition module, balance module, CAN and power module; Described malfunction coefficient module is made up of seven segmentation charactrons; Described data acquisition module is made up of voltage monitoring chip LTC6802, NTC thermistor and buffer circuit; Described balance module is made up of matrix switch, flyback DC/DC transducer and balance controller; Described CAN receives and dispatches TJA1050 by high-speed CAN, two-channel digital isolating chip IS07221A is formed; Described power module is made up of voltage stabilizing PWB2405CS and IB00505S; Described central management unit CMU comprises: data memory module, current detection module, RS485 communication module, CAN and power module, and described data memory module is made up of Micro SD card and level transferring chip ADG3308; Current detection module is made up of Hall current sensor, operational amplifier; Described RS485 communication module is made up of differential data transponder chip SN65LBC184D; Described CAN is by high-speed CAN transceiver TJA1050, and two-channel digital isolating chip IS07221A is formed; Described power module is made up of voltage stabilizing chip PWA2412MD, K7805, REG1117 and IB00505S; Described host computer comprises: HMI touch-screen.
2. a kind of energy storage battery management system based on degree of depth learning network according to claim 1, is characterized in that, the state-of-charge SOC of battery of the present invention and accumulator health status SOH estimates by deep neural network and learning algorithm realization thereof; Described deep neural network construction step is: the first step, determine input layer and output layer neuron: using battery total capacity, voltage, electric current, temperature and the SOC in k-1 moment as the input layer of neural network, using the output layer neuron of SOC and SOH of battery as neural network; Second step, determines activation function, the neural network hidden layer number of plies and hidden layer neuron number, thus sets up neural network; 3rd step, adjustment network parameter, trains the neural network created, and estimates SOC and SOH.
CN201410169036.2A 2014-04-25 2014-04-25 Energy storage battery management system based on deep learning network Withdrawn CN104881001A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410169036.2A CN104881001A (en) 2014-04-25 2014-04-25 Energy storage battery management system based on deep learning network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410169036.2A CN104881001A (en) 2014-04-25 2014-04-25 Energy storage battery management system based on deep learning network

Publications (1)

Publication Number Publication Date
CN104881001A true CN104881001A (en) 2015-09-02

Family

ID=53948531

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410169036.2A Withdrawn CN104881001A (en) 2014-04-25 2014-04-25 Energy storage battery management system based on deep learning network

Country Status (1)

Country Link
CN (1) CN104881001A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105137242A (en) * 2015-09-09 2015-12-09 南京航空航天大学 Single-phase photovoltaic inverter on-line state monitoring and residual life prediction method
CN105634058A (en) * 2016-01-22 2016-06-01 广东志成冠军集团有限公司 Intelligent balancing method and intelligent balancing system for battery pack
CN107887658A (en) * 2016-09-29 2018-04-06 法乐第(北京)网络科技有限公司 Accumulator cell assembly, motor vehicle
CN109784480A (en) * 2019-01-17 2019-05-21 武汉大学 A kind of power system state estimation method based on convolutional neural networks
CN110412470A (en) * 2019-04-22 2019-11-05 上海博强微电子有限公司 Electric automobile power battery SOC estimation method
CN110531281A (en) * 2019-09-09 2019-12-03 合肥工业大学 The method and system of health status for estimated driving force secondary battery unit
CN110658460A (en) * 2019-09-29 2020-01-07 东软睿驰汽车技术(沈阳)有限公司 Battery life prediction method and device for battery pack
CN112254815A (en) * 2020-08-28 2021-01-22 华电电力科学研究院有限公司 Electrochemical energy storage station temperature monitoring system based on three-dimensional infrared imaging temperature measurement
CN112557908A (en) * 2020-12-17 2021-03-26 温州大学 SOC and SOH joint estimation method for lithium ion power battery
US11171498B2 (en) 2017-11-20 2021-11-09 The Trustees Of Columbia University In The City Of New York Neural-network state-of-charge estimation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2444511A (en) * 2006-12-06 2008-06-11 Iti Scotland Ltd Neural network battery management
CN102385660A (en) * 2010-09-03 2012-03-21 湘潭大学 Evaluation method for system on chip (SOC) of charging station battery
CN102738525A (en) * 2012-06-14 2012-10-17 沈阳中科正方新能源技术有限公司 Battery management system of vehicle-mounted lithium power battery
CN103501026A (en) * 2013-09-17 2014-01-08 武汉大学 Lithium battery management system based on Labview
CN103558554A (en) * 2013-10-15 2014-02-05 广东电网公司电力科学研究院 On-line estimating method for SOH of new energy automobile power battery

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2444511A (en) * 2006-12-06 2008-06-11 Iti Scotland Ltd Neural network battery management
CN102385660A (en) * 2010-09-03 2012-03-21 湘潭大学 Evaluation method for system on chip (SOC) of charging station battery
CN102738525A (en) * 2012-06-14 2012-10-17 沈阳中科正方新能源技术有限公司 Battery management system of vehicle-mounted lithium power battery
CN103501026A (en) * 2013-09-17 2014-01-08 武汉大学 Lithium battery management system based on Labview
CN103558554A (en) * 2013-10-15 2014-02-05 广东电网公司电力科学研究院 On-line estimating method for SOH of new energy automobile power battery

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105137242A (en) * 2015-09-09 2015-12-09 南京航空航天大学 Single-phase photovoltaic inverter on-line state monitoring and residual life prediction method
CN105137242B (en) * 2015-09-09 2018-04-13 南京航空航天大学 Single-phase photovoltaic inverter on-line condition monitoring and method for predicting residual useful life
CN105634058A (en) * 2016-01-22 2016-06-01 广东志成冠军集团有限公司 Intelligent balancing method and intelligent balancing system for battery pack
CN105634058B (en) * 2016-01-22 2018-05-25 广东志成冠军集团有限公司 A kind of intelligent equalization method of battery pack and intelligent equalization system
CN107887658A (en) * 2016-09-29 2018-04-06 法乐第(北京)网络科技有限公司 Accumulator cell assembly, motor vehicle
US11171498B2 (en) 2017-11-20 2021-11-09 The Trustees Of Columbia University In The City Of New York Neural-network state-of-charge estimation
CN109784480A (en) * 2019-01-17 2019-05-21 武汉大学 A kind of power system state estimation method based on convolutional neural networks
CN110412470A (en) * 2019-04-22 2019-11-05 上海博强微电子有限公司 Electric automobile power battery SOC estimation method
CN110412470B (en) * 2019-04-22 2021-09-21 上海博强微电子有限公司 SOC estimation method for power battery of electric vehicle
CN110531281B (en) * 2019-09-09 2021-08-13 合肥工业大学 Method and system for estimating state of health of power storage battery unit
CN110531281A (en) * 2019-09-09 2019-12-03 合肥工业大学 The method and system of health status for estimated driving force secondary battery unit
CN110658460A (en) * 2019-09-29 2020-01-07 东软睿驰汽车技术(沈阳)有限公司 Battery life prediction method and device for battery pack
CN112254815A (en) * 2020-08-28 2021-01-22 华电电力科学研究院有限公司 Electrochemical energy storage station temperature monitoring system based on three-dimensional infrared imaging temperature measurement
CN112557908A (en) * 2020-12-17 2021-03-26 温州大学 SOC and SOH joint estimation method for lithium ion power battery

Similar Documents

Publication Publication Date Title
CN104881001A (en) Energy storage battery management system based on deep learning network
CN103675686B (en) Electric vehicle power battery charging and discharging operating condition simulation system and method
JP6111275B2 (en) Battery control device
CN102231546A (en) Battery management system with balanced charge and discharge functions and control method thereof
CN103208828B (en) A kind of series-connected cell group management system
CN103563210B (en) System for storing power comprising modularized BMS connection structure and method for controlling same
US20140111164A1 (en) Battery control device and battery system
JP6101714B2 (en) Battery control device, battery system
CN201781037U (en) Battery management system of electric automobile
EP2720054A1 (en) Battery control device and battery system
CN101595620A (en) Supply of electric power control method, the control program of power-supply system, power-supply system and record the computer-readable recording medium of control program
CN104242393A (en) Battery management system based on dynamic SOC estimation system
JP6364127B2 (en) Failure diagnosis apparatus and failure diagnosis method for storage battery array
CN103091633A (en) Estimating device and method of lead-acid storage battery level
US8030894B2 (en) System and method for rebalancing a vehicle battery
CN105471020A (en) Circuit and method for detection of battery cell drain
CN106505267A (en) A kind of full-sized car fuel cell management system and method
CN106199450A (en) A kind of battery health evaluation system and method
Ma et al. Fuzzy logic based power and thermal management system design for multi-cell lithium-ion battery bank protection and operation
CN103296324B (en) Vehicle power battery pack charging method
CN110190347A (en) A kind of lithium battery management system applied to communication base station
CN104348206A (en) Distributed battery management device and dual-redundancy power supply system
CN201945674U (en) Recorder for battery discharge capacity
CN205508969U (en) Lead -acid batteries management system
CN107546433A (en) Battery information processing method and processing device

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
EXSB Decision made by sipo to initiate substantive examination
WW01 Invention patent application withdrawn after publication

Application publication date: 20150902

WW01 Invention patent application withdrawn after publication