CN117595462B - Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack - Google Patents

Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack Download PDF

Info

Publication number
CN117595462B
CN117595462B CN202410069994.6A CN202410069994A CN117595462B CN 117595462 B CN117595462 B CN 117595462B CN 202410069994 A CN202410069994 A CN 202410069994A CN 117595462 B CN117595462 B CN 117595462B
Authority
CN
China
Prior art keywords
battery
data
charge
discharge
battery pack
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
Application number
CN202410069994.6A
Other languages
Chinese (zh)
Other versions
CN117595462A (en
Inventor
欧煌
王宏建
张军
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.)
Suzhou Sidaoer New Energy Technology Co ltd
Original Assignee
Suzhou Sidaoer New 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 Suzhou Sidaoer New Energy Technology Co ltd filed Critical Suzhou Sidaoer New Energy Technology Co ltd
Priority to CN202410069994.6A priority Critical patent/CN117595462B/en
Publication of CN117595462A publication Critical patent/CN117595462A/en
Application granted granted Critical
Publication of CN117595462B publication Critical patent/CN117595462B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
    • H01M10/00—Secondary cells; Manufacture thereof
    • H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
    • H01M10/44—Methods for charging or discharging
    • H01M10/441—Methods for charging or discharging for several batteries or cells simultaneously or sequentially
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/392—Determining battery ageing or deterioration, e.g. state of health
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
    • 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
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/30—Circuit design
    • G06F30/32—Circuit design at the digital level
    • G06F30/33—Design verification, e.g. functional simulation or model checking
    • G06F30/3323—Design verification, e.g. functional simulation or model checking using formal methods, e.g. equivalence checking or property checking
    • 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
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • 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
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0029—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with safety or protection devices or circuits
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0029—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with safety or protection devices or circuits
    • H02J7/00302—Overcharge protection
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0029—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with safety or protection devices or circuits
    • H02J7/00306—Overdischarge protection
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
    • H02J7/0048—Detection of remaining charge capacity or state of charge [SOC]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
    • H02J7/005—Detection of state of health [SOH]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/007—Regulation of charging or discharging current or voltage
    • H02J7/00712—Regulation of charging or discharging current or voltage the cycle being controlled or terminated in response to electric parameters
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Power Engineering (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Electrochemistry (AREA)
  • Manufacturing & Machinery (AREA)
  • Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • General Chemical & Material Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Secondary Cells (AREA)

Abstract

The invention discloses a self-adaptive charge and discharge control system of a secondary energy storage battery pack and a control method thereof, wherein the self-adaptive charge and discharge control system comprises the following steps: the system comprises a data acquisition module, a prediction model module, an optimization scheduling module and a control execution module; the control method comprises the following steps: s1, collecting a battery pack topological structure, physical parameters of each battery, historical charge and discharge data, historical SOC data and historical SOH data; s2, establishing an equivalent circuit unit of each battery and integrating a battery pack equivalent circuit model; s3, performing time sequence analysis through an ARIMA model, and continuously predicting the SOC data and the SOH data of each battery in a short period; s4, outputting a prediction result confidence interval, adjusting constraint variables of each equivalent circuit unit during charging and discharging, and outputting a primary scheduling scheme; and S5, monitoring the battery state, and optimizing constraint variables according to the preliminary scheduling scheme and the equivalent circuit model of the battery pack. The invention can dynamically adjust according to the actual condition of the battery pack, and improves the charge and discharge efficiency and the safety of the battery pack.

Description

Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack
Technical Field
The invention relates to the technical field of electric energy storage systems, in particular to a device for charging or depolarizing a battery pack or supplying power to a load by the battery pack, and particularly relates to an adaptive charging and discharging control system of a secondary energy storage battery pack and a control method thereof.
Background
The secondary energy storage battery is a battery capable of being used for multiple charge and discharge cycles, and is also called a rechargeable battery. The reversibility of chemical reaction is utilized, and the active substance can be activated in a charging mode after the battery is discharged for continuous use. Common secondary batteries are lead storage batteries, nickel-hydrogen batteries, nickel-cadmium batteries, lithium ion batteries, polymer lithium ion batteries, and the like. These batteries have higher energy and power densities, which can provide longer service times and faster charge speeds.
In the secondary battery, active materials are carriers of chemical reactions, which are chemically reacted during charge and discharge, thereby achieving storage and release of energy. The charge-discharge cycle of the rechargeable battery can reach thousands to tens of thousands times, so that the rechargeable battery is more economical and practical compared with a dry battery.
The self-discharge of a secondary battery is also called charge retention capability, which refers to the retention capability of the amount of electricity stored in the battery under certain environmental conditions in an open state. In general, self-discharge is mainly affected by the manufacturing process, materials, and storage conditions, and is one of the main parameters for measuring the performance of a battery.
In the prior art, the charge-discharge real-time state of the battery pack can be monitored only according to a monitoring system so as to control the charge-discharge current and the charge-discharge voltage of the battery, and the battery pack cannot be regulated and controlled independently according to each battery in the battery pack, so that the overall regulation and control precision is reduced; and the state change of the battery pack in a period of time in the future cannot be predicted, so that the charge and discharge of the battery pack are unbalanced, and proper measures cannot be taken in advance to optimize the performance of the battery pack and prolong the service life of the battery pack.
Disclosure of Invention
The invention overcomes the defects of the prior art and provides a self-adaptive charge and discharge control system and a control method of a secondary energy storage battery pack.
In order to achieve the above purpose, the invention adopts the following technical scheme: the control method of the self-adaptive charge and discharge control system of the secondary energy storage battery pack is characterized by comprising the following steps of:
S1, data acquisition and analysis: data acquisition is carried out on the topological structure of the battery pack, the physical parameters and the historical charge and discharge data of each battery in the battery pack, the historical SOC data and the historical SOH data;
S2, building a battery pack model: establishing an equivalent circuit unit of each battery according to the topological structure and the physical parameters acquired in the step S1, integrating the equivalent circuit units of a plurality of batteries, and establishing an equivalent circuit model of the battery pack;
S3, establishing an interval prediction model: carrying out time sequence analysis on the historical charge and discharge data, the historical SOC data and the historical SOH data in the S1 through an ARIMA model, and continuously carrying out short-term prediction on the SOC data and the SOH data of each battery;
s4, adjusting battery scheduling: outputting a confidence interval of the prediction result in real time, and synchronously adjusting constraint variables in the equivalent circuit units of each battery during charging and discharging by the system according to the confidence interval to output a primary scheduling scheme;
S5, optimizing battery scheduling: and (3) monitoring state parameters of each battery in real time, and optimizing constraint variables according to a preliminary scheduling scheme and an equivalent circuit model of the battery pack to realize dynamic scheduling of the charging and discharging processes of a plurality of batteries.
In a preferred embodiment of the present invention, in the step S2, the establishing an equivalent circuit model of the battery pack specifically includes the following steps:
s21, establishing an equivalent circuit unit of a single battery by using a Thevenin model;
S22, determining the connection relation between each equivalent circuit unit according to the topological structure of the battery pack, wherein the connection relation is one or more of series connection and parallel connection, and a preliminary equivalent circuit model is formed;
S23, assigning a value to the preliminary equivalent circuit model of each battery according to the physical parameters of each battery, setting experimental data, recording actual measurement data through discharge performance test, and predicting and recording prediction data by using the same experimental data;
And S24, comparing the measured data with the predicted data, verifying the accuracy of the preliminary equivalent circuit model, and optimizing the model when the deviation exists to obtain a final equivalent circuit model of the battery pack.
In a preferred embodiment of the present invention, in the step S3, training is further performed on the historical charge-discharge policy data, the historical SOC data, and the historical SOH data by using a long-short-term memory network model, so as to obtain a more accurate prediction result.
In a preferred embodiment of the present invention, in the step S1, the topology of the battery pack includes: the number of batteries in the battery pack and the connection mode of each battery; the physical parameters of each cell in the battery include: battery type, battery capacity, and battery internal resistance; the historical charge and discharge data includes: average duration of charge and discharge, frequency of charge and discharge, maximum limit of charge and discharge current, charge termination voltage and discharge termination voltage.
In a preferred embodiment of the present invention, in the step S1, collecting the charge-discharge temperature data, the charge-discharge rate data and the cycle number data of each battery in the battery pack is further included.
In a preferred embodiment of the present invention, in the step S3, the method further includes accelerating a model prediction process, and specifically includes the following steps:
s31, sequentially performing abnormal value removal, missing value filling and smooth data processing on historical charge and discharge strategy data, SOC data and SOH data, and reducing data dimension;
s32, establishing an ARIMA model according to the data obtained in the S31, and optimizing the ARIMA model by adopting maximum likelihood estimation and a distributed computing technology to reduce model quantity;
s33, parallel computing technology is utilized to process multiple groups of data in parallel, and the model prediction process is quickened.
The invention also provides a self-adaptive charge and discharge control system of the secondary energy storage battery pack, which comprises the following components: the system comprises a data acquisition module, a prediction model module, an optimization scheduling module and a control execution module;
the data acquisition module comprises: the parameter acquisition unit is used for acquiring the topological structure of the battery pack, the physical parameters and the historical charge and discharge data of each battery in the battery pack, the historical SOC data and the historical SOH data; a data storage unit that stores the collected data as history data; the data preprocessing unit is used for cleaning and standardizing the historical data;
The predictive model module includes: the monomer model building unit is used for building an equivalent circuit unit of each battery according to the preprocessed data; the combined model unit is used for integrating the monomer model into an equivalent circuit model; prediction unit: carrying out short-term prediction on the SOC data and the SOH data of each battery according to the preprocessed data to generate a confidence interval;
The optimal scheduling module comprises: the scheduling algorithm unit is used for generating a preliminary scheduling scheme of the battery according to the confidence interval; the optimization algorithm unit is used for optimizing the preliminary scheduling scheme according to the equivalent circuit model of the battery pack and the state parameters monitored in real time;
the control execution module includes: the control instruction generation unit is used for generating a specific scheduling instruction according to the optimization result; and the execution control unit is used for sending the scheduling instruction to the battery management system for execution.
In a preferred embodiment of the present invention, the data acquisition module further includes a load acquisition unit, configured to acquire a load demand power parameter and a load limiting voltage parameter.
In a preferred embodiment of the present invention, the prediction model module further includes: the parameter modeling unit is used for establishing a dynamic parameter model of the battery pack such as a battery capacity attenuation model and the like; and the state modeling unit is used for establishing a battery state quantity model and a battery SOH state quantity model.
In a preferred embodiment of the present invention, the system further includes a communication interface module, where the communication interface module includes: the sensing execution unit is used for directly interacting with the battery pack through the sensor and the actuator to acquire data in real time and sending the data to the data preprocessing unit; and the CAN bus unit is used for linking all the modules through a CAN bus to realize real-time data transmission.
The invention solves the defects existing in the background technology, and has the following beneficial effects:
(1) The invention provides a self-adaptive charge and discharge control system of a secondary energy storage battery pack and a control method thereof.
(2) When the equivalent circuit model of the battery pack is built, the equivalent circuit unit of the single battery is built by using the Thevenin model, the model can describe the internal characteristics of the battery more accurately, and the connection mode among the batteries in the battery pack can be described more accurately according to the topological structure and the physical parameters of the battery pack, so that an accurate basis is provided for the building of the model; and then verifying the accuracy of the model through assignment, and ensuring the reliability and the accuracy of the model.
(3) According to the invention, the charge and discharge temperature data, the charge and discharge rate data and the cycle number data of each battery in the battery pack are acquired, other factors influencing the state of the battery are added, and the performance and the state of the battery pack can be more comprehensively known by comprehensively considering various factors, so that more accurate data support is provided for subsequent charge and discharge control.
(4) According to the invention, the long-term and short-term memory network model is used for training the historical charge-discharge strategy data, the historical SOC data and the historical SOH data, the data are continuously changed in the battery pack, and the LSTM model is used for better capturing the changes, so that the robustness of the system is enhanced, and the system can be stably and accurately predicted in the face of various changes.
(5) The invention reduces the data dimension and the calculated amount and the storage requirement by processing the data; and the ARIMA model is optimized by combining maximum likelihood estimation with a distributed computing technology, so that model quantity is reduced, and a plurality of groups of data are processed in parallel by utilizing a parallel computing technology, so that the model prediction process is quickened. The method is beneficial to improving the real-time performance and response speed of the system, and the charge and discharge control of the battery pack is more flexible and efficient.
(6) According to the invention, the constraint variables of charge or discharge are adjusted by predicting the SOC and the SOH, the SOC and the SOH can reflect the electric quantity level and the health condition of the battery, the state of the battery can be better known by real-time monitoring, the overcharge and the overdischarge are avoided, and the battery is protected from being damaged; meanwhile, the optimal charging time and charging current can be determined, the charging efficiency is improved, and the energy loss in the charging process is reduced; and the performance of the battery can be predicted, so that adjustment measures are adopted in advance, and the influence degree of the performance reduction of the battery on the whole system is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art;
FIG. 1 is a schematic diagram of an adaptive charge and discharge control system according to the present invention;
fig. 2 is a flowchart of a control method of the adaptive charge and discharge control system in the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, an adaptive charge and discharge control system of a secondary energy storage battery pack includes: the system comprises a data acquisition module, a prediction model module, an optimization scheduling module and a control execution module.
The data acquisition module comprises:
The parameter acquisition unit is used for collecting key data from the secondary energy storage battery pack;
a data storage unit for storing the collected data as history data;
And the data preprocessing unit is used for cleaning and standardizing the historical data.
In the parameter acquisition unit, the key data includes: the topology of the battery pack, the physical parameters and historical charge and discharge data of each battery in the battery pack, historical SOC data, and historical SOH data.
The topology of the battery pack includes: the number of the batteries in the battery pack and the connection mode of each battery.
The physical parameters of each cell in the battery include: battery type, rated battery capacity, and battery internal resistance.
The historical charge and discharge data includes: average duration of charge and discharge, frequency of charge and discharge, maximum limit of charge and discharge current, charge termination voltage and discharge termination voltage.
SOC: state of Charge, an abbreviation for battery remaining capacity, which represents the ratio of energy available in the battery to total energy, is an important indicator for measuring the battery Charge level.
SOH: state of Health, an abbreviation for battery State of Health, which indicates the State of deterioration and service life of a battery, is used to evaluate the performance and maintenance requirements of the battery.
Historical SOC data: SOC refers to the state of charge of a battery, is a key technology of a battery management system, and is a quantization index of the remaining capacity of an electric vehicle. The historical SOC data refers to the state of charge change condition of the battery in the past period of time, and is used for evaluating the charge and discharge performance of the battery, predicting the residual electric quantity of the battery and optimizing the operation strategy of the battery management system.
Historical SOH data: SOH refers to the state of health of a battery, which reflects the performance and life of the battery. Historical SOH data refers to the state of health change of a battery over a period of time, used to evaluate the performance and life of the battery, predict the replacement time of the battery, and optimize the operating strategy of the battery management system.
The parameter acquisition unit comprises:
sensor network: the method is used for monitoring key parameters such as temperature, voltage, current and the like of the battery pack.
A data acquisition card: for collecting data from the sensor network and transmitting it to the data storage unit.
The data storage unit includes:
mass storage device: a hard disk or SSD is used to store the collected historical data.
Database management system: for managing, querying and retrieving stored historical data.
The data preprocessing unit includes:
Data washer: for flushing data, including deleting duplicate data, processing missing values, identifying and processing outliers.
A data converter: for converting the raw data to meet the requirements of subsequent analysis or model training.
The data acquisition module further comprises a load acquisition unit used for acquiring load demand electric quantity parameters and load limiting voltage parameters.
The load acquisition unit includes:
load current sensor: the system is used for monitoring the current demand of the load in real time, and ensuring that the system can not exceed the supply capacity of the power supply while meeting the load demand.
Load voltage sensor: the voltage limiting device is used for monitoring the voltage limit of the load, and ensuring that the system can not damage the load while meeting the voltage requirement of the load.
Load data acquisition card: and receiving and processing the measurement data of the load current sensor and the load voltage sensor, and transmitting the data to a data storage unit.
The charge and discharge process of the battery pack can be controlled more accurately by collecting the load demand electric quantity parameters, and overcharge or undercharge of the battery can be avoided, so that stable operation of the battery is ensured, and the service life of the battery is prolonged. The system can adjust the output voltage of the battery pack according to the load demand and the limiting voltage by collecting the load limiting voltage parameter so as to meet the operation demand of the equipment and ensure the safe operation of the battery pack. For example, when the load demand is high, the system may select a battery with higher charge/discharge efficiency or adjust the output voltage of the battery pack to meet the load demand.
The predictive model module includes:
The monomer model building unit is used for building an equivalent circuit unit of each battery according to the preprocessed data;
the combined model unit is used for integrating the monomer model into an equivalent circuit model;
prediction unit: and carrying out short-term prediction on the SOC data and the SOH data of each battery according to the preprocessed data to generate a confidence interval.
The monomer model building unit is used for building an equivalent circuit model for each battery according to the preprocessed data. The equivalent circuit model is a mathematical model that describes the electrical behavior and performance of the battery. It simulates the internal structure and performance of a battery by a circuit consisting of a resistor and a capacitor.
The monomer model building unit includes:
A data receiver receiving and processing battery data from the preprocessing unit; based on the received battery data, an equivalent circuit model is built for each battery through an equivalent circuit model algorithm;
The combined model unit is used for integrating the monomer models into an integral equivalent circuit model. The combined model unit finds the optimal model parameters and the combined mode through a specific algorithm, wherein the specific algorithm can be one of an optimization algorithm or a system integration method.
The prediction unit is used for carrying out short-term prediction on the SOC and SOH data of each battery according to the preprocessed data and the established equivalent circuit model. The prediction unit may learn a rule from the history data using one of a statistical learning algorithm or a machine learning algorithm, and predict a future trend. The prediction results will generate a confidence interval that indicates the accuracy and confidence level of the prediction.
Data after pretreatment: refers to data after data cleaning and standardization processing, and the data processing methods can improve the accuracy and the usability of the data.
Equivalent circuit model: a mathematical model for describing the electrical behavior and performance of a battery converts the internal structure and performance of the battery into a circuit model consisting of resistive and capacitive electronic components.
Monomer model: an equivalent circuit model is built for each cell that reflects the electrical characteristics and performance of the individual cell.
Combining the model: the plurality of cell models are integrated into an integral equivalent circuit model, which can describe the electrical behavior and performance of the whole battery pack more accurately.
Confidence interval: indicating the extent of accuracy and confidence in the predicted outcome.
The prediction model module further comprises: the parameter modeling unit is used for establishing a dynamic parameter model of the battery pack; and the state modeling unit is used for establishing a battery state quantity model and a battery SOH state quantity model.
The parameter modeling unit is used for establishing a dynamic parameter model of the battery pack and providing basic data and model support for subsequent battery state prediction and performance evaluation.
The state modeling unit is used for establishing an SOC state quantity model and an SOH state quantity model of the battery and providing accurate state quantity representation for subsequent battery state prediction.
And a dynamic parameter model of the battery pack, such as a battery capacity attenuation model, is established through the parameter modeling unit, so that the charge and discharge capacity of the battery pack can be predicted, a basis is provided for optimizing the scheduling, and the stable operation of the battery pack is ensured. The state modeling unit can help the system to better know the health state of the battery and predict the performance degradation of the battery, so that corresponding maintenance measures are adopted in advance, and the safe operation of the battery pack is ensured.
The optimal scheduling module comprises:
The scheduling algorithm unit is used for generating a preliminary scheduling scheme of the battery according to the confidence interval, taking various factors into consideration, including the current state, the historical use condition and the environmental condition of the battery, and preparing the most suitable scheduling scheme through a scheduling algorithm; wherein, the scheduling algorithm can be one of a genetic algorithm or a particle swarm algorithm;
The optimization algorithm unit is used for optimizing the preliminary scheduling scheme through an optimization algorithm according to the equivalent circuit model of the battery pack and the state parameters of each battery monitored in real time; wherein the optimization algorithm may be one of gradient descent or simulated annealing.
The state parameters of each cell include: real-time SOC data of each battery, real-time voltage of each battery, real-time temperature of each battery, total voltage of a battery pack and total current of the battery pack.
The control execution module includes:
The control instruction generation unit is used for generating a specific executable scheduling instruction according to the output result of the optimization algorithm so as to be executed by the battery management system; the control instruction generation unit includes: the optimization result receiving sub-module is in charge of receiving the optimization result output from the optimization algorithm unit; the instruction coding submodule converts the optimization result into an instruction format which can be identified and executed by the battery management system; an instruction verification sub-module that ensures that the generated instructions are correct and secure before being sent to the battery management system;
The execution control unit is used for sending the scheduling instruction to the battery management system for execution; the execution control unit includes: the communication interface sub-module is responsible for establishing communication connection with the battery management system and ensuring accurate transmission of instructions; the instruction sending sub-module establishes connection with the battery management system through the communication interface and sends the verified scheduling instruction to the battery management system; and the feedback receiving sub-module receives the execution feedback from the battery management system and ensures that the instruction is correctly received and executed.
The system also comprises a communication interface module, wherein the communication interface module comprises:
the sensing execution unit is directly interacted with the battery pack, and is used for collecting data of the battery pack in real time through the sensor and the actuator and sending the data to the data preprocessing unit;
And the CAN bus unit is used for linking all the modules through a CAN bus to realize real-time data transmission.
CAN bus is a communication protocol used in automotive and other industrial applications. In the battery management system, stability and instantaneity of data communication between the modules CAN be ensured by using the CAN bus.
The state of the battery can be better known through real-time monitoring, overcharge and overdischarge are avoided, and the battery is protected from damage; meanwhile, the optimal charging time and charging current can be determined, the charging efficiency is improved, and the energy loss in the charging process is reduced; and the performance of the battery can be predicted, so that adjustment measures are adopted in advance, and the influence degree of the performance reduction of the battery on the whole system is reduced.
As shown in fig. 2, the invention further provides a control method of the adaptive charge-discharge control system of the secondary energy storage battery pack, which comprises the following steps:
S1, data acquisition and analysis: data acquisition is carried out on the topological structure of the battery pack, the physical parameters and the historical charge and discharge data of each battery in the battery pack, the historical SOC data and the historical SOH data;
S2, building a battery pack model: establishing an equivalent circuit unit of each battery according to the topological structure and the physical parameters acquired in the step S1, integrating the equivalent circuit units of a plurality of batteries, and establishing an equivalent circuit model of the battery pack;
S3, establishing an interval prediction model: carrying out time sequence analysis on the historical charge and discharge data, the historical SOC data and the historical SOH data in the S1 through an ARIMA model, and continuously carrying out short-term prediction on the SOC data and the SOH data of each battery;
s4, adjusting battery scheduling: outputting a confidence interval of the prediction result in real time, and synchronously adjusting constraint variables in the equivalent circuit units of each battery during charging and discharging by the system according to the confidence interval to output a primary scheduling scheme;
S5, optimizing battery scheduling: and (3) monitoring state parameters of each battery in real time, and optimizing constraint variables according to a preliminary scheduling scheme and an equivalent circuit model of the battery pack to realize dynamic scheduling of the charging and discharging processes of a plurality of batteries.
In step S1, the topology of the battery pack includes: the number of batteries in the battery pack and the connection mode of each battery; the physical parameters of each cell in the battery include: battery type, rated battery capacity, and battery internal resistance; the historical charge and discharge data includes: average duration of charge and discharge, frequency of charge and discharge, maximum limit of charge and discharge current, charge termination voltage and discharge termination voltage.
The specific steps for collecting the topological structure of the battery pack are as follows: and determining the number and the connection mode of the batteries in the battery pack, identifying the series connection and parallel connection relation among the batteries through a battery connection identifier, and recording the overall topological structure of the battery pack. The battery connection identifier can automatically identify the connection relation between batteries and record the topological structure.
In step S1, collecting charge-discharge temperature data, charge-discharge rate data and cycle number data of each battery in the battery pack; other factors influencing the battery state are added, and the performance and the state of the battery pack can be more comprehensively known by comprehensively considering various factors, so that more accurate data support is provided for subsequent charge and discharge control.
For example, data collection and analysis based on internet of things technology:
Firstly, data acquisition is carried out, an Internet of things sensor and an actuator are arranged for a battery pack, and the equipment can monitor physical parameters such as the topological structure of the battery pack, the voltage, the current and the temperature of each battery in real time; the data acquired by the sensor and the actuator are transmitted to a cloud server in real time through the internet of things technology;
When the data is stored, a cloud storage service is used, the collected data is stored in a cloud database, and the safety and accessibility of the data are ensured; abnormal values, noise data and the like are identified and removed through an algorithm, so that the accuracy of the data is ensured; carrying out standardized processing on data of different dimensions and ranges so as to facilitate subsequent analysis; and automatically classifying and marking the historical charge and discharge data, the historical SOC data and the historical SOH data by using a machine learning algorithm.
For example, local server-based data collection and analysis:
A local server is configured at the place where the battery pack is located, and the server is provided with a high-performance data acquisition card, so that the topological structure of the battery pack and the physical parameters of each battery can be acquired in real time; transmitting the acquired data to a database of a local server in a wired connection mode; a database is established on a local server and used for storing collected data; to ensure data security, data is backed up to external storage devices periodically.
Batch processing is carried out on the collected data at regular intervals, and the batch processing comprises the steps of data cleaning, data arrangement and the like; according to actual requirements, setting a custom data preprocessing rule to meet analysis requirements in a specific scene; manually classifying and marking historical charge and discharge data, historical SOC data and historical SOH data according to experience by professionals; and combining experience of professionals and a machine learning algorithm, semi-automatically classifying the data, and improving classification accuracy and efficiency.
In step S2, establishing an equivalent circuit model of the battery pack specifically includes the steps of:
s21, establishing an equivalent circuit unit of a single battery by using a Thevenin model; according to the characteristics and parameters of the battery, abstracting the battery into an equivalent circuit of a voltage source and an internal resistor; the voltage source represents the electromotive force of the battery, and the internal resistance represents the internal resistance of the battery;
S22, determining the connection relation between each equivalent circuit unit according to the topological structure of the battery pack, wherein the connection relation is serial connection, parallel connection or series-parallel connection, and a preliminary equivalent circuit model is formed;
S23, assigning a value to the preliminary equivalent circuit model of each battery according to the physical parameters of each battery, setting experimental data, recording actual measurement data through discharge performance test, and predicting and recording prediction data by using the same experimental data;
And S24, comparing the measured data with the predicted data, verifying the accuracy of the preliminary equivalent circuit model, and optimizing the model when the deviation exists to obtain a final equivalent circuit model of the battery pack.
The internal characteristics of the batteries can be described more accurately by using the Thevenin model, and the connection mode among the batteries in the battery pack can be described more accurately according to the topological structure and the physical parameters of the battery pack, so that an accurate basis is provided for the establishment of the model; and then verifying the accuracy of the model through assignment, and ensuring the reliability and the accuracy of the model.
For example, in step S21, the electromotive force of one cell is 3.7V, and the internal resistance is 0.1 Ω. According to the Thevenin model, the equivalent circuit unit of the battery is represented as a voltage source of 3.7V and an internal resistance of 0.1 Ω. This completes the establishment of the equivalent circuit unit of the single battery.
For example, in step S22, there is a battery pack including 3 lithium ion batteries, in which two batteries are connected in series and the other battery is separate. And determining the connection relation between each equivalent circuit unit according to the topological structure of the battery pack to form a preliminary equivalent circuit model.
The method comprises the following specific steps: first, two cells connected in series are regarded as a whole, and are called a battery pack 1. The equivalent circuit model of the battery pack 1 can be expressed as a series circuit of one electromotive force source and one internal resistance. The electromotive force source voltage was assumed to be 6V and the internal resistance was assumed to be 0.2Ω.
The individual cells are then regarded as a battery pack 2. The equivalent circuit model of the battery pack 2 can be expressed as a series circuit of one electromotive force source and one internal resistance. The electromotive force source voltage was assumed to be 3.7V, and the internal resistance was assumed to be 0.1 Ω.
And finally, connecting the battery pack 1 and the battery pack 2 in parallel to form a preliminary equivalent circuit model of the whole battery pack. I.e. the sources of electromotive forces of the battery 1 and the battery 2 are directly connected, while their internal resistances are also directly connected. Thus, a preliminary equivalent circuit model of a battery pack including 3 lithium ion batteries was obtained.
In step S23, the experimental data includes: charging current and discharging current, charging and discharging duration, experimental environment temperature, and battery residual capacity; the prediction data includes: predicting a voltage change curve, and predicting a current change curve of electric quantity discharged in a discharging process; the measured data includes: and actually measuring a voltage change curve, and actually measuring a current change curve by using the discharged electric quantity recorded when the discharge is finished.
For example, the specific embodiment of step S23 is: the rated capacity of one battery in the battery pack is 2Ah, the rated voltage is 3.7V, and the internal resistance is 0.1 Ω. Recording actual discharge data of the battery by using a discharge performance test, performing time series analysis on the actual measurement data by using an ARIMA model, and predicting future discharge behavior of the battery.
Experimental data included:
Charging current and discharging current: the charge current was 2A and the discharge current was 3A.
Duration of charge and discharge: the charge duration was 2 hours and the discharge duration was 1 hour.
Experimental ambient temperature: 25 ℃.
Battery remaining capacity: the remaining battery capacity at the beginning of the experiment was 80%.
The prediction data includes:
Predicting a voltage change curve: the ARIMA model predicts a voltage change curve in the future discharging process according to the historical voltage data, the sampling interval of the battery in the historical discharging process is known to be 5s, the voltage is reduced from 4.2V to 3.7V after the battery is continuously discharged for 20min, and the ARIMA model predicts that the voltage of the battery is continuously reduced from 3.7V to 3.1V under the same discharging condition.
Electric quantity discharged in the discharging process: the ARIMA model predicts the electric quantity discharged in the future discharging process according to the historical electric quantity data, the sampling interval of the battery in the historical discharging process is 5s, the electric quantity of 1Ah is discharged after continuous discharging for 20min, and the ARIMA model predicts that the battery can continuously discharge the electric quantity of 1Ah under the same discharging condition.
Predicting a current change curve: the ARIMA model predicts a current change curve in a future discharging process according to historical current data, the sampling interval of the battery in the historical discharging process is known to be 5s, after the battery is continuously discharged for 20min, the current rises from 2A to 2.5A, and the ARIMA model predicts that the current of the battery can continuously rise from 2.5A to 3A under the same discharging condition.
The measured data includes:
in the experimental process, a voltage collector is used for recording the voltage change condition of the battery in real time. The voltage was recorded to gradually decrease from 3.75V to 3.13V during discharge.
The discharged electric quantity recorded at the end of discharge: at the end of the discharge, the actual amount of discharged electricity was recorded as 1.2Ah.
Measured current change curve: during the experiment, the current change condition of the battery is recorded in real time by using a current collector, and the current is gradually increased from 2.57A to 3.12A in the discharging process.
Then, comparison is performed in step S24, for example:
Voltage change curve comparison: the ARIMA model predicts a voltage drop from 3.7V to 3.0V. The measured data shows a voltage drop from 3.7V to 3.1V. The two have a deviation of 0.1V.
And (3) discharging electric quantity comparison:
the ARIMA model predicts that 1Ah electric quantity is discharged, and the measured data show that 1.2Ah electric quantity is discharged, and the two have deviation of 0.2 Ah.
Comparison of current change curves: the ARIMA model predicts a current rise from 2.5A to 3A, and the measured data shows a current rise from 2.5A to 3.1A, with a consistent trend of current change, but a deviation of 0.1A.
And optimizing the sampling interval from 5s to 1s according to the deviation, adding a battery internal resistance parameter into the model, retraining an optimized ARIMA equivalent circuit model for the battery internal resistance R=50mΩ, and predicting newly acquired data.
The comparison with the new round of measured data shows that: the voltage predicted value is adjusted from original 3.7V-3.1V to 3.743V-3.121V, and the matching degree with the actually measured 3.75V-3.13V is improved; the current predicted value is adjusted from 2.5A-3A to 2.545A-3.109A, and the matching degree with the actually measured 2.57A-3.12A is improved; the discharge capacity prediction is adjusted to be 1.199Ah from 1Ah, and the matching degree with the actually measured 1.2Ah is improved. Through repeated optimization, a battery pack equivalent circuit model with the precision within 0.001 units can be finally established, and the high-precision requirement is met.
In step S3, the method further comprises training the historical charge-discharge strategy data, the historical SOC data and the historical SOH data by using the long-short-term memory network model, wherein the data are continuously changed in the battery pack, and the LSTM model can be used for better capturing the changes, so that the robustness of the system is enhanced, and the system can be stably and accurately predicted in the face of various changes.
In step S3, the method further includes an acceleration model prediction process, specifically including the following steps:
s31, removing abnormal values from historical charge-discharge strategy data, SOC data and SOH data, filling the missing values by using linear interpolation or spline interpolation, and then performing smooth data processing by using average filtering or exponential smoothing to reduce the data dimension;
For example, the raw data contains SOC and SOH data per hour of 3 batteries, the data amount being 100 time points; check if there are outliers in the data, such as SOC > 100% or SOH < 0. Battery 1 SOC data at time points 10 and 30 were detected as 120% and-5%, respectively, and considered outliers and the missing values were filled in using linear interpolation. For battery 1, the 10 th time points were linearly interpolated using the 9 th and 11 th time points; the 30 th time point was linearly interpolated using the 29 th and 31 th time points.
The SOC and SOH data for each battery are then smoothed using average filtering, respectively: data for battery 1 SOC: new data point= (original data point + previous data point + next data point)/3.
The next step is to reduce the data dimension: the data at each 5 time points were averaged to obtain new data of length 20. For example: new data point 1= (original data point 1+.+ original data point 5)/5.
S32, establishing an ARIMA model according to the data obtained in the S31, determining the model order, and optimizing the ARIMA model by adopting maximum likelihood estimation and a distributed computing technology to reduce the model quantity;
for example, using the time-series data of length 5 obtained after the processing of S31, an ARIMA model is constructed to obtain:
battery 1 SOC: ARIMA (1, 0, 1), battery 1 SOH: ARIMA (0, 1);
battery 2 SOC: ARIMA (0, 1, 2), battery 2 SOH: ARIMA (1, 0);
battery 3 SOC: ARIMA (2, 1), battery 3 SOH: ARIMA (1, 0, 2).
S33, parallel computing technology is utilized to process multiple groups of data in parallel, sample data to be predicted are distributed to multiple nodes, each node independently performs prediction computation, and the model prediction process is quickened.
By processing the data, the data dimension is reduced, and the calculated amount and the storage requirement are reduced; and the ARIMA model is optimized by combining maximum likelihood estimation with a distributed computing technology, so that model quantity is reduced, and a plurality of groups of data are processed in parallel by utilizing a parallel computing technology, so that the model prediction process is quickened. The method is beneficial to improving the real-time performance and response speed of the system, and the charge and discharge control of the battery pack is more flexible and efficient.
In step S4, constraint variables include: a charge-discharge current limit for each battery, a total charge-discharge current limit for the battery, a charge-discharge voltage limit for each battery, and a total voltage limit for the battery.
In step S4, the system synchronously adjusts constraint variables in the equivalent circuit unit of each battery during charging and discharging according to the confidence interval, and specifically includes the following steps:
S41, adjusting a charge-discharge current limit value: according to the confidence interval of the SOC data and the SOH data of each battery, the charge and discharge current limit value of each battery is adjusted; counting error distribution of an actual measurement value and a predicted value in a period of time, and determining a standard error range as a judging standard of a low-high confidence interval;
S42, according to the SOC predicted value, the charge and discharge current limit value of each battery is adjusted to be +/-5-10% in a high confidence interval or a low confidence interval. If the predicted value is in the high confidence interval, the charge-discharge current limit value and the charge-discharge voltage limit value are increased, and if the predicted value is in the low confidence interval, the charge-discharge current limit value and the charge-discharge voltage limit value are reduced.
For example, there is a battery pack containing 3 cells numbered 1,2, 3; the charge-discharge current limit of each battery is adjusted according to the confidence interval of the SOC data and SOH data of each battery.
Firstly, counting error distribution of an actual measurement value and a predicted value in a period of time to obtain the following data:
the SOC predicted value of battery 1 is 80%, the measured value is 82%, and the standard error is 1%;
the SOC predicted value of battery 2 was 70%, the measured value was 68%, and the standard error was 2%;
the SOC prediction value of battery 3 was 90%, the actual measurement value was 88%, and the standard error was 1.5%.
From the above data we can get confidence intervals for each cell:
The confidence interval of cell 1 is [81%, 83% ];
the confidence interval of battery 2 is [66%, 70% ];
the confidence interval for cell 3 was [86.5%, 91.5% ].
The adjustment amplitude of the charge-discharge current limit value of each battery is set to + -5-10%.
The SOC prediction value of battery 1 was 82.5% in the high confidence interval [81%, 83% ]. The charge-discharge current limit and the charge-discharge voltage limit of the battery 1 need to be increased according to step S41. The limit of the charge and discharge current of the battery 1 is 10A, the limit of the charge and discharge voltage is 30V, the limit of the charge and discharge current of the battery 1 is adjusted to 11-11.5A according to the adjustment range of + -5-10%, and the limit of the charge and discharge voltage is adjusted to 31.5-33V.
The SOC prediction value of battery 2 was 67.3% within the low confidence interval [66%, 70% ]. The charge-discharge current limit and the charge-discharge voltage limit of the battery 2 need to be lowered according to step S41. The limit value of the charge and discharge current of the battery 2 is 8A, the limit value of the charge and discharge voltage is 25V, the limit value of the charge and discharge current of the battery 2 is 7.2-7.6A according to the adjustment range of + -5-10%, and the limit value of the charge and discharge voltage is 23.8-24.5V.
For battery 3, the SOC prediction value was 89.6% within the high confidence interval [86.5%, 91.5% ]. According to step S41, the charge-discharge current limit and the charge-discharge voltage limit of the battery 3 need to be raised. The limit value of the charge and discharge current of the battery 3 is 15A, and the limit value of the charge and discharge voltage is 40V, so that the limit value of the charge and discharge current of the battery 3 is 15.75-16.5A, and the limit value of the charge and discharge voltage is 42-44V according to the adjustment range of + -5-10%.
In step S5, the state parameters of each battery include: real-time SOC data of each battery, real-time voltage of each battery, real-time temperature of each battery, total voltage of a battery pack and total current of the battery pack.
In one embodiment, the following steps may be employed:
S51, establishing a connection topological graph of battery nodes and edges according to the series connection and parallel connection relation of the battery packs;
S52, setting state variables including voltage V, current I, SOC and SOH for each node in the topological graph; setting the maximum allowable current for each side as a flow constraint;
S53, mapping the charge and discharge tasks in the preliminary scheduling scheme to flow tasks in the topological graph; applying a minimum charge flow model, regarding a charge-discharge task as flow, regarding a battery state variable constraint as a capacity constraint of a side, and regarding an objective function as a minimum charge-discharge amount min f,minf= Σci*Qi, wherein c i is a battery capacity loss amount of battery i, and Q i is a charge-discharge amount of battery i.
S54, solving the objective function and the constraint condition by using an interior point method or a simple x method, modeling the objective function and the constraint condition by using a matrix representation method, calling a linear programming solver, and solving an optimal flow distribution variable value to obtain an optimal flow distribution scheme meeting the topological structure and the preliminary task; according to the optimal flow scheme, regulating a state variable constraint range of each battery, wherein the state variable constraint range comprises an SOC range and upper and lower current limits;
and S55, monitoring and evaluating the actual state of each battery in real time. And further optimizing the variable constraint of each battery according to the evaluation result.
S56, repeatedly executing the steps S52-S55, continuously tightening and optimizing the variable constraint of each battery on the basis of meeting the topology limit, and simultaneously considering the residual capacity of the battery pack to prevent the premature termination of the task caused by overuse of certain batteries. When needed, the flow tasks are re-planned, and the battery scheduling is adjusted to improve efficiency and prolong service life.
For example, one battery pack comprising 3 series-connected batteries is optimized, numbered 1,2, 3.
Establishing a battery connection topology diagram, wherein the battery connection topology diagram comprises 3 nodes representing 3 batteries and 2 sides connecting the 3 batteries; setting a state variable initial value for each node:
cell 1: soc=0.8, v=3.7v, i=0a, t=25 ℃;
cell 2: soc=0.7, v=3.7v, i=0a, t=25 ℃;
cell 3: soc=0.6, v=3.7v, i=0a, t=25 ℃.
A maximum current limit is set for each side, I max = 5A.
The preliminary scheduling task is to charge the battery 1 and the battery 3, and the charge amount is 1Ah. Mapping it to traffic tasks, battery 1 flows into 1A, and battery 3 flows into 1A. Then, a minimum cost flow model is established: the objective function is min (c 1Q1 + c3Q3), where c1=c3=0.01 (loss 0.01 watt-hour per ampere hour). And (3) applying linear programming solution to obtain optimal flow distribution meeting the topological structure and the preliminary task, wherein the minimum charge and discharge quantity min f =0.02A.
The battery state was monitored in real time and the battery 2 was found to be 35 ℃ higher in temperature. The model is further optimized according to the temperature rise condition of the battery 2, and the maximum current limit I max,Imax =3a of the battery 2 is reduced. And repeatedly executing optimization to obtain a final scheduling scheme.
The conditions that the traffic tasks need to be re-planned and the battery scheduling needs to be adjusted to improve the charge and discharge efficiency and prolong the service life of the battery include:
The battery state changes significantly, for example, some battery SOCs drop too fast, so that subsequent tasks cannot be completed;
The task demand is changed, and the original planning cannot meet the new task demand;
battery parameters change, for example, certain battery capacities are severely reduced, and scheduling efficiency is affected;
The topology of the battery pack changes, for example, a battery drops off or a new battery is added;
the operation monitoring data shows that the battery utilization rate and balance of the original scheduling scheme are not optimized;
After long-time operation, the battery state accumulation effect is obvious, and the original scheduling scheme cannot ensure even use of the battery;
The environmental condition or the system load changes, and the original scheme is difficult to adapt to the new situation;
The optimization effect of the algorithm is continuously improved, and the latest optimization scheme is necessary to replace the original scheme;
periodically re-optimizing to adapt to the dynamic change characteristics of the battery performance and parameters.
The above-described preferred embodiments according to the present invention are intended to suggest that, from the above description, various changes and modifications can be made by the person skilled in the art without departing from the scope of the technical idea of the present invention. The technical scope of the present invention is not limited to the description, but must be determined according to the scope of claims.

Claims (10)

1. The control method of the self-adaptive charge and discharge control system of the secondary energy storage battery pack is characterized by comprising the following steps of:
S1, data acquisition and analysis: data acquisition is carried out on the topological structure of the battery pack, the physical parameters and the historical charge and discharge data of each battery in the battery pack, the historical SOC data and the historical SOH data;
S2, building a battery pack model: establishing an equivalent circuit unit of each battery according to the topological structure and the physical parameters acquired in the step S1, integrating the equivalent circuit units of a plurality of batteries, and establishing an equivalent circuit model of the battery pack;
S3, establishing an interval prediction model: carrying out time sequence analysis on the historical charge and discharge data, the historical SOC data and the historical SOH data in the S1 through an ARIMA model, and continuously carrying out short-term prediction on the SOC data and the SOH data of each battery;
s4, adjusting battery scheduling: outputting a confidence interval of the prediction result in real time, and synchronously adjusting constraint variables in the equivalent circuit units of each battery during charging and discharging by the system according to the confidence interval to output a primary scheduling scheme; wherein the constraint variables include: a charge-discharge current limit for each battery, a total charge-discharge current limit for the battery, a charge-discharge voltage limit for each battery, and a total voltage limit for the battery;
the system synchronously adjusts constraint variables in the equivalent circuit unit of each battery during charging and discharging according to the confidence interval, and specifically comprises the following steps:
Adjusting the limit value of the charge and discharge current: according to the confidence interval of the SOC data and the SOH data of each battery, the charge and discharge current limit value of each battery is adjusted; counting error distribution of an actual measurement value and a predicted value in a period of time, and determining a standard error range as a judging standard of a low-high confidence interval;
according to the state of charge (SOC) predicted value being in a high confidence interval or a low confidence interval, the adjustment amplitude of the charge-discharge current limit value of each battery is +/-5-10%; if the predicted value is in the high confidence interval, the charge-discharge current limit value and the charge-discharge voltage limit value are improved, and if the predicted value is in the low confidence interval, the charge-discharge current limit value and the charge-discharge voltage limit value are reduced;
S5, optimizing battery scheduling: the state parameters of each battery are monitored in real time, constraint variables are optimized according to a preliminary scheduling scheme and an equivalent circuit model of the battery pack, and dynamic scheduling of the charging and discharging processes of a plurality of batteries is achieved;
Wherein the state parameters of each battery include: real-time SOC data of each battery, real-time voltage of each battery, real-time temperature of each battery, total voltage of a battery pack and total current of the battery pack;
the dynamic scheduling of the charging and discharging processes of the batteries specifically comprises the following steps:
S51, establishing a connection topological graph of battery nodes and edges according to the series connection and parallel connection relation of the battery packs;
S52, setting state variables including voltage V, current I, SOC and SOH for each node in the topological graph; setting the maximum allowable current for each side as a flow constraint;
S53, mapping the charge and discharge tasks in the preliminary scheduling scheme to flow tasks in the topological graph; applying a minimum charge flow model, regarding a charge-discharge task as flow, regarding a battery state variable constraint as a capacity constraint of a side, and regarding an objective function as a minimum charge-discharge amount min f,minf＝Σci*Qi, wherein c i is a battery capacity loss amount of a battery i, and Q i is a charge-discharge amount of the battery i;
S54, solving the objective function and the constraint condition by using an interior point method or a simple x method, modeling the objective function and the constraint condition by using a matrix representation method, calling a linear programming solver, and solving an optimal flow distribution variable value to obtain an optimal flow distribution scheme meeting the topological structure and the preliminary task; according to the optimal flow scheme, regulating a state variable constraint range of each battery, wherein the state variable constraint range comprises an SOC range and upper and lower current limits;
s55, monitoring and evaluating the actual state of each battery in real time, and further optimizing the variable constraint of each battery according to the evaluation result;
s56, repeatedly executing the steps S52-S55, continuously tightening and optimizing the variable constraint of each battery on the basis of meeting the topology limit, and simultaneously considering the residual capacity of the battery pack to prevent the premature termination of the task caused by overuse of certain batteries.
2. The control method of the adaptive charge and discharge control system of the secondary energy storage battery pack according to claim 1, wherein the control method comprises the following steps: in the step S2, the establishment of the equivalent circuit model of the battery pack specifically includes the following steps:
s21, establishing an equivalent circuit unit of a single battery by using a Thevenin model;
S22, determining the connection relation between each equivalent circuit unit according to the topological structure of the battery pack, wherein the connection relation is one or more of series connection and parallel connection, and a preliminary equivalent circuit model is formed;
S23, assigning a value to the preliminary equivalent circuit model of each battery according to the physical parameters of each battery, setting experimental data, recording actual measurement data through discharge performance test, and predicting and recording prediction data by using the same experimental data;
And S24, comparing the measured data with the predicted data, verifying the accuracy of the preliminary equivalent circuit model, and optimizing the model when the deviation exists to obtain a final equivalent circuit model of the battery pack.
3. The control method of the adaptive charge and discharge control system of the secondary energy storage battery pack according to claim 1, wherein the control method comprises the following steps: in the step S3, training is further performed on the historical charge-discharge strategy data, the historical SOC data and the historical SOH data by using the long-short-period memory network model, so as to obtain a more accurate prediction result.
4. The control method of the adaptive charge and discharge control system of the secondary energy storage battery pack according to claim 1, wherein the control method comprises the following steps: in the S1, the topology of the battery pack includes: the number of batteries in the battery pack and the connection mode of each battery; the physical parameters of each cell in the battery include: battery type, battery capacity, and battery internal resistance; the historical charge and discharge data includes: average duration of charge and discharge, frequency of charge and discharge, maximum limit of charge and discharge current, charge termination voltage and discharge termination voltage.
5. The control method of the adaptive charge and discharge control system of the secondary energy storage battery pack according to claim 1, wherein the control method comprises the following steps: in the step S1, collecting charge and discharge temperature data, charge and discharge rate data and cycle number data of each battery in the battery pack is further included.
6. The control method of the adaptive charge and discharge control system of the secondary energy storage battery pack according to claim 1, wherein the control method comprises the following steps: in the step S3, the method further includes an acceleration model prediction process, specifically including the following steps:
s31, sequentially performing abnormal value removal, missing value filling and smooth data processing on historical charge and discharge strategy data, SOC data and SOH data, and reducing data dimension;
s32, establishing an ARIMA model according to the data obtained in the S31, and optimizing the ARIMA model by adopting maximum likelihood estimation and a distributed computing technology to reduce model quantity;
s33, parallel computing technology is utilized to process multiple groups of data in parallel, and the model prediction process is quickened.
7. An adaptive charge and discharge control system for a secondary energy storage battery, comprising: the system comprises a data acquisition module, a prediction model module, an optimization scheduling module and a control execution module, and is characterized in that,
The data acquisition module comprises: the parameter acquisition unit is used for acquiring the topological structure of the battery pack, the physical parameters and the historical charge and discharge data of each battery in the battery pack, the historical SOC data and the historical SOH data; a data storage unit that stores the collected data as history data; the data preprocessing unit is used for cleaning and standardizing the historical data;
The predictive model module includes: the monomer model building unit is used for building an equivalent circuit unit of each battery according to the preprocessed data; the combined model unit is used for integrating the monomer model into an equivalent circuit model; prediction unit: carrying out short-term prediction on the SOC data and the SOH data of each battery according to the preprocessed data to generate a confidence interval;
The optimal scheduling module comprises: the scheduling algorithm unit is used for generating a preliminary scheduling scheme of the battery according to the confidence interval, and specifically comprises the following steps: synchronously adjusting constraint variables in an equivalent circuit unit of each battery during charging and discharging according to a confidence interval, wherein the constraint variables comprise a charging and discharging current limit value of each battery, a total charging and discharging current limit value of a battery pack, a charging and discharging voltage limit value of each battery and a total voltage limit value of the battery pack, and generating a preliminary scheduling scheme of the batteries; the optimization algorithm unit is used for optimizing the preliminary scheduling scheme according to the equivalent circuit model of the battery pack and the state parameters monitored in real time, and specifically comprises the following steps: establishing a connection topology diagram of battery nodes and sides according to the monitored real-time SOC data of each battery, the real-time voltage of each battery, the real-time temperature of each battery, the total voltage of the battery pack and the total current of the battery pack and the serial and parallel connection relation of the battery pack; setting state variables including voltage V, current I, SOC and SOH for each node in the topological graph; setting the maximum allowable current for each side as a flow constraint; mapping the charge and discharge tasks in the preliminary scheduling scheme to flow tasks in the topological graph; applying a minimum charge flow model, regarding a charge-discharge task as flow, regarding a battery state variable constraint as a capacity constraint of a side, and regarding an objective function as a minimum charge-discharge amount min f,minf＝Σci*Qi, wherein c i is a battery capacity loss amount of a battery i, and Q i is a charge-discharge amount of the battery i; solving the objective function and the constraint condition by using an interior point method or a simple x method, modeling the objective function and the constraint condition by using a matrix representation method, calling a linear programming solver, and solving an optimal flow distribution variable value to obtain an optimal flow distribution scheme meeting the topological structure and the preliminary task; according to the optimal flow scheme, regulating a state variable constraint range of each battery, wherein the state variable constraint range comprises an SOC range and upper and lower current limits; monitoring and evaluating the actual state of each battery in real time, and further optimizing the variable constraint of each battery according to the evaluation result;
the control execution module includes: the control instruction generation unit is used for generating a specific scheduling instruction according to the optimization result; and the execution control unit is used for sending the scheduling instruction to the battery management system for execution.
8. The adaptive charge and discharge control system of a secondary energy storage battery of claim 7, wherein: the data acquisition module also comprises a load acquisition unit used for acquiring load demand electric quantity parameters and load limiting voltage parameters.
9. The adaptive charge and discharge control system of a secondary energy storage battery of claim 7, wherein: the prediction model module further comprises: the parameter modeling unit is used for establishing a dynamic parameter model of the battery pack; and the state modeling unit is used for establishing a battery state quantity model and a battery SOH state quantity model.
10. The adaptive charge and discharge control system of a secondary energy storage battery of claim 7, wherein: the system also comprises a communication interface module, wherein the communication interface module comprises: the sensing execution unit is used for directly interacting with the battery pack through the sensor and the actuator to acquire data in real time and sending the data to the data preprocessing unit; and the CAN bus unit is used for linking all the modules through a CAN bus to realize real-time data transmission.
CN202410069994.6A 2024-01-18 2024-01-18 Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack Active CN117595462B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410069994.6A CN117595462B (en) 2024-01-18 2024-01-18 Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410069994.6A CN117595462B (en) 2024-01-18 2024-01-18 Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack

Publications (2)

Publication Number Publication Date
CN117595462A CN117595462A (en) 2024-02-23
CN117595462B true CN117595462B (en) 2024-05-17

Family

ID=89911867

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410069994.6A Active CN117595462B (en) 2024-01-18 2024-01-18 Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack

Country Status (1)

Country Link
CN (1) CN117595462B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112782591A (en) * 2021-03-22 2021-05-11 浙江大学 Lithium battery SOH long-term prediction method based on multi-battery data fusion
CN117318255A (en) * 2023-11-30 2023-12-29 北京中铁建电气化设计研究院有限公司 Battery state analysis system and method based on big data visualization

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112782591A (en) * 2021-03-22 2021-05-11 浙江大学 Lithium battery SOH long-term prediction method based on multi-battery data fusion
CN117318255A (en) * 2023-11-30 2023-12-29 北京中铁建电气化设计研究院有限公司 Battery state analysis system and method based on big data visualization

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王东.机会约束的两阶段能源管理系统全局优化策略.《制造业自动化》.第第45卷卷(第第9期期),第56-63页. *

Also Published As

Publication number Publication date
CN117595462A (en) 2024-02-23

Similar Documents

Publication Publication Date Title
CN110568361B (en) Method for predicting health state of power battery
CN107957562B (en) Online prediction method for residual life of lithium ion battery
KR102335296B1 (en) Wireless Network based Battery Management System
JP7497432B2 (en) Battery performance prediction
CN111190111B (en) Electrochemical energy storage battery state of charge estimation method, device and system
CN109100655B (en) Data processing method and device for power battery
WO2023185601A1 (en) Method and device for determining state of health information of battery, and battery system
CN110850298A (en) Lithium battery SOH estimation method and system based on data driving
US11835589B2 (en) Method and apparatus for machine-individual improvement of the lifetime of a battery in a battery-operated machine
CN110658459A (en) Lithium ion battery state of charge estimation method based on bidirectional cyclic neural network
CN116609676B (en) Method and system for monitoring state of hybrid energy storage battery based on big data processing
CN115114878A (en) Method and device for online prediction of battery life of energy storage power station and storage medium
CN110806540B (en) Battery cell test data processing method, device and system and storage medium
CN115616425A (en) Battery performance analysis method, electronic equipment and energy storage system
CN117317408A (en) Battery and heat optimization management method based on big data and artificial intelligence
CN115219904A (en) Method and device for operating a system for providing an electrochemical cell stack model
CN110850315A (en) Method and device for estimating state of charge of battery
CN117054892B (en) Evaluation method, device and management method for battery state of energy storage power station
CN117595462B (en) Self-adaptive charge and discharge control system and control method of secondary energy storage battery pack
CN117330961A (en) Battery pack monitoring method and device, electronic equipment and readable storage medium
Avadhanula et al. Comparative study of mathematical models and data driven models for battery performance parameter estimation
CN115980612A (en) Satellite battery pack health state assessment method, system and equipment
CN115951225A (en) Battery equalization optimization capacity estimation method and device
CN115483763A (en) Lead-acid battery energy storage power station monitoring management system and method
CN115219932A (en) Method and device for evaluating the relative aging state of a battery of a device

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