理想汽车BMS AI算法工程师-北京/上海
社招全职5年以上汽车研发地点:北京状态:招聘
任职要求
1.硕士及以上学历,自动化、计算机、电化学等相关专业,5年以上BMS算法开发经验, 2.精通优化算法(如EKF/UKF、最小二乘)与AI算法(XGBoost/LSTM/强化学习); 3.能融合电化学模型(等效电路、单粒子模型)与数据驱动方法解决SOC/SOH/SOP估计问题; 4.熟练使用Python/MATLAB开发算法,熟悉MBD流程(Simulink建模、HIL测试); 5.深入理解电池热-电化学耦合模型、老化模型及析锂预测方法,具备多物理场仿真能力如COMSOL; 6.主导过至少1个量产项目的SOX算法开发,熟悉车规级开发流程(ISO 26262/ASPICE)。
工作职责
1.SOX算法开发与全生命周期管理 1.1 SOC高精度估计:基于电化学机理模型(如DFN模型)与数据驱动方法(LSTM、Transformer),融合电池电压、电流、温度、内阻等多维数据,设计自适应卡尔曼滤波算法(如UKF、AEKF),解决低温/高倍率工况下的累积误差问题(目标误差<1%); 构建动态参数辨识框架(如基于遗传算法或粒子群优化),实时校准电池容量、内阻等关键参数,提升SOC估算的长期稳定性; 1.2 SOH预测与退化建模:利用迁移学习技术,将实验室加速老化数据泛化至实际车载场景,构建基于容量衰减、内阻增长、SEI膜演化的多维度退化模型,实现SOH误差<2%; 1.3 SOP动态边界计算:基于电芯温度、SOC、老化状态的实时反馈,建立多约束条件下的峰值功率预测模型(如电化学-热耦合模型),防止过充/过放风险,支持极端工况(如赛道模式)下的动态功率调整。 2. AI模型开发与优化 2.1开发轻量化神经网络模型(如MobileNet、TinyML架构),通过剪枝、量化、知识蒸馏等技术将模型压缩至嵌入式平台可运行规模,满足实时性要求(响应延迟<50ms); 2.2数据驱动与仿真验证: 构建电池全生命周期数据库(覆盖电芯、模组、系统层级),通过SQL处理TB级数据,提取关键特征(如充放电曲线拐点、弛豫电压特性)用于模型训练;使用MATLAB/Simulink搭建多物理场耦合仿真平台(电化学+热力学+机械应力),验证算法在极端工况下的鲁棒性,并通过HIL测试实现算法闭环迭代; 2.3 与BMS硬件团队协作,优化AI算法在嵌入式平台的资源分配,支持C代码自动生成与功能安全认证。
包括英文材料
学历+
算法+
https://roadmap.sh/datastructures-and-algorithms
Step by step guide to learn Data Structures and Algorithms in 2025
https://www.hellointerview.com/learn/code
A visual guide to the most important patterns and approaches for the coding interview.
https://www.w3schools.com/dsa/
XGBoost+
[英文] What is XGBoost?
https://www.ibm.com/think/topics/xgboost
XGBoost (eXtreme Gradient Boosting) is a distributed, open-source machine learning library that uses gradient boosted decision trees, a supervised learning boosting algorithm that makes use of gradient descent.
https://www.youtube.com/watch?v=BJXt-WdeJJo
takes a deep dive into one of the most powerful machine learning algorithm, eXtreme Gradient Boosting, using a Jupyter notebook with Python.
LSTM+
https://colah.github.io/posts/2015-08-Understanding-LSTMs/
Humans don’t start their thinking from scratch every second.
https://d2l.ai/chapter_recurrent-modern/lstm.html
The term “long short-term memory” comes from the following intuition.
https://developer.nvidia.com/discover/lstm
A Long short-term memory (LSTM) is a type of Recurrent Neural Network specially designed to prevent the neural network output for a given input from either decaying or exploding as it cycles through the feedback loops.
https://www.youtube.com/watch?v=YCzL96nL7j0
Basic recurrent neural networks are great, because they can handle different amounts of sequential data, but even relatively small sequences of data can make them difficult to train.
强化学习+
https://cloud.google.com/discover/what-is-reinforcement-learning?hl=en
Reinforcement learning (RL) is a type of machine learning where an "agent" learns optimal behavior through interaction with its environment.
https://huggingface.co/learn/deep-rl-course/unit0/introduction
This course will teach you about Deep Reinforcement Learning from beginner to expert. It’s completely free and open-source!
https://www.kaggle.com/learn/intro-to-game-ai-and-reinforcement-learning
Build your own video game bots, using classic and cutting-edge algorithms.
SOC+
https://www.arm.com/resources/education/books/modern-soc
The aim of this textbook is to expose aspiring and practising SoC designers to the fundamentals and latest developments in SoC design and technologies using examples of Arm Cortex-A technology and related IP blocks and interfaces.
https://www.arm.com/resources/education/education-kits/introduction-to-soc
To produce students with solid introductory knowledge on the basics of SoC design and key practical skills required to implement a simple SoC on an FPGA and write embedded programs targeted at the microprocessor to control the peripherals.
https://www.youtube.com/watch?v=dokgLSAhqHI
A key part of the digital innovation revolution has been the embrace of the SoC, or system-on-chip.
Python+
https://liaoxuefeng.com/books/python/introduction/index.html
中文,免费,零起点,完整示例,基于最新的Python 3版本。
https://www.learnpython.org/
a free interactive Python tutorial for people who want to learn Python, fast.
https://www.youtube.com/watch?v=K5KVEU3aaeQ
Master Python from scratch 🚀 No fluff—just clear, practical coding skills to kickstart your journey!
https://www.youtube.com/watch?v=rfscVS0vtbw
This course will give you a full introduction into all of the core concepts in python.
MATLAB+
https://matlabacademy.mathworks.com/?page=1&sort=featured
Learn MATLAB and Simulink through interactive, in-product exercises
https://www.mathworks.com/help/matlab/getting-started-with-matlab.html
Millions of engineers and scientists worldwide use MATLAB® to analyze and design the systems and products transforming our world.
https://www.youtube.com/watch?v=7f50sQYjNRA
Learn the fundametnals of MATLAB in this tutorial for engineers, scientists, and students.
Simulink+
https://www.mathworks.com/help/simulink/getting-started-with-simulink.html
Simulink® is a block diagram environment for multidomain simulation and Model-Based Design.
相关职位
校招电池开发
1. 电池高精度状态估计算法开发; 2. 电池高精度预测算法开发; 3. 高精度电池模型开发与应用; 4. 基于算法开发流程完成项目交付; 5. 针对市场问题或云端大数据,能够快速高效定位并能制定优化方案。
校招电池开发
1. 电池高精度状态估计算法开发; 2. 电池高精度预测算法开发; 3. 高精度电池模型开发与应用; 4. 基于算法开发流程完成项目交付; 5. 针对市场问题或云端大数据,能够快速高效定位并能制定优化方案。
社招3年以上研发技术类
1. 综合能源管理系统建模与优化: 设计光储充、微电网、厂区节能改造等综合能源场景的能源调度策略,实现多能源多负荷(光、储、热等)协同优化。 研究开发电站发电、用电预测的核心算法,结合电力市场化交易、动态电价等关键因素,设计最优运行控制策略算法,以实现能源经济效益的最大化;同时提升电力系统的稳定性和运行效率。 2. 算法研发与落地: 主导机器学习/深度学习模型(时序预测、运筹优化、强化学习)的研发,应用于能源供需预测、风险评估、功率控制等场景。 推动算法工程化,完成从数据预处理、模型训练到部署上线的全流程闭环。 3. 前沿技术探索: 跟踪生成式AI、大模型在能源领域的应用,结合业务需求进行竞品分析和技术迭代。 4. 跨团队协作: 与运营、产品团队协作,将算法嵌入能源管理平台(EMS),支撑智慧能源运营决策。
更新于 2025-07-28