Remote Management Solutions for Lithium Batteries

User Requirements:

For lithium battery dealers or users with a large number of lithium battery products, it is essential to be able to control and manage their lithium battery products easily and quickly, saving a lot of manpower and material resources. The current "Bluetooth/WiFi+APP" solution can only achieve point-to-point management. It is not possible to achieve batch management, and the management or user needs to be within a certain distance of the battery in order to operate it, so it cannot be managed from a distance!

Technical Difficulties:

1. Battery data acquisition:

To understand and manage lithium batteries, it is necessary to master all the core data of lithium batteries. At present, some Bluetooth and WiFi communication methods can only grasp the real-time voltage, current, voltage difference and other basic data of lithium batteries. However, these data are not enough to make decisions on lithium battery management, but also need the history of lithium battery operation data, fault alarm data, etc.

2. Transmission of information

When we have data on lithium batteries, we need to be able to view it online at any time by means of appropriate data transmission. Especially when dealing with large quantities of battery data remotely, in addition to the stable transmission of real-time status data, we also need to know the battery's historical operation data, which requires extremely high stability and speed of data transmission! This is something that cannot be achieved with the "Bluetooth/WiFi + Mobile App" solution.

3. Storage of data

How to store and read the data of a large volume of Li-ion batteries in real time places higher demands on the way the data is stored. Many Li-ion batteries will use localised storage of battery data, i.e. a local storage module is integrated into the BMS. The disadvantages of this solution are also obvious: due to limited space, once the local storage space is full, it will automatically overwrite the previous data; in addition, the data between individual batteries are independent of each other and cannot maximise the effect of the data.

4. Data analysis

As we mentioned above, lithium batteries generate a large amount of data, including the real-time status of the lithium battery, operating history, fault alarms and other key data, which is decisive for users to make decisions on battery management. How to make good use of these data is an important reflection of the level of intelligence of the lithium battery management platform. At present there are almost no management platforms with this capability.

5. Remote Management

Most of the lithium batteries are installed in the car, so the location is not fixed, so you can manage them remotely regardless of geographical location to better meet the actual needs of customers, in addition to batch operation of lithium batteries, new/retracted sub-accounts, set account permissions and other functions.

To achieve a truly powerful remote management platform, we need to solve several technical difficulties mentioned above! Today we would like to introduce you to LYBATT's PaaS (Platform as a Service) based battery cloud management platform + smart hardware remote management solution.

Battery Cloud

The PaaS-based battery cloud management platform is a powerful platform with intelligent AI algorithm models and machine learning technology; users can access rich battery management capabilities and data service capabilities through the remote monitoring platform on the computer side to meet the business needs of different users.

Smart hardware:

• 4G network card: provides data transmission in most parts of the world
• Data sending and receiving terminals: sending real-time battery data to the Battery Cloud data platform, as well as receiving control commands from the management terminal, downloading updated data packages and other functions;
• Intelligent BMS: compatible with the Battery Cloud data platform for management.

4G Remote Management Platform Features

It is based on the strong technical strength, LYBATT's 4G remote management platform can realize 4 major sections and 20+ rich functions.

• Collaborative management: Remote balancing, SOC correction, SOP control.
• Safety alert: Abnormal voltage difference, Thermal field monitoring, Cell deterioration identification, Failure detection.
• BMS connection: Link management, Protocol Codec, Data upload, Downlink Instructions Secure transmission, Load balancing, Traffic optimization.
• Data services: Shadow device mapping, Real time data push, Battery operation report, BMS fault analysis, Battery performance evaluation, Health status assessment, Exclusive customization.

Collaborative control in the cloud

The AI algorithm-based model engine, Leigh, uses the near-infinite storage capacity and computing power of cloud servers to build big data analysis models based on machine learning methods for non-smoothly varying, multi-dimensional massive operational data, continuously iterating to optimise battery management parameters and achieve collaborative control in the cloud.

Cell imbalance life cycle< 20mV@SOC Interval 20%-80%, algorithm is particularly effective for LFP battery, increasing single range by 10%

Utilizing individual/common data, Cloud-based charging/discharging collaborative management as well as cell power control technology extend the service life of lithium-ion battery by 15%

Multi-timescale based dQ/dv differential curve feature extraction technique and cumulative capacity difference cloud correction technique, The SOC/SOH error is less than 3%

Proactive safety management

Based on the powerful functions of the intelligent hardware, it can achieve real-time monitoring and analysis of battery status change trends, identify application security risks in advance, and provide 7*24-hour all-round security protection. When the intelligent BMS sends out the battery warning information, it is sent to Battery Could through the data sending and receiving platform, and then the platform will send out the warning content through the remote monitoring platform on the computer side, mobile phone, telephone, SMS and email to remind customers to make the correct response in time.

• Abnormal Voltage Difference Alert: Based on the single charge/discharge curve and capacity decay curve, we use Isolation Forrest anomaly detection algorithm through machine learning to detect abnormal voltage difference and send alert messages.
• Abnormal Internal Resistance Alert: The internal resistance and current of the battery are determined by comparing the difference between the charge cut-off voltage and the open circuit voltage. The charging internal resistance is predicted and analyzed using a deep learning recurrent neural network (RNN).
• Temperature Anomaly Alert: Based on BP neural network model, we improve its error calculation method and weight adjustment method to predict the temperature curve of cells, packs and MOS's temperature in real time to achieve abnormal alerting.
• Cell Consistency Alert: Real-time detection of cell voltage, temperature, capacity, SOC and other parameters, through the establishment of a consistency evaluation model to score cell performance, timely maintenance or replacement of abnormal cells.
• Remaining Life Alert: The impact factor of RUL prediction is constructed by using battery charging and discharging parameters. The relationship between constant voltage drop discharge time series and capacity is measured by using correlation analysis method. The time series prediction is implemented based on Gaussian process regression (GPR) algorithm to estimate the RUL of batteries.

Summary

LYBATT's 4G remote management platform also provides mature and standardized platform interfaces, such as account system, battery binding management, event notification engine, OTA management, timed task engine, etc., to meet the individual business needs of different types of customers.