Early Thermal Runaway Detection & Condition Monitoring in Traction Battery Packs through Gas Detection

该项目提出了一种利用单个气体传感器监测牵引电池组内锂离子电池的新方法。电池组中的每个电池单元都将被涂上一种代理材料，这种材料对每个电池单元的分解特征略有不同。这些材料将被设计成在非常特定的温度下“激活”和释放气体。通过检测释放气体的成分，单个传感器可以检测和定位电池组内任何电池发生的热事件。此外，锂离子电池中可能发生的极其危险的热失控机制可以通过电池温度直接表征。这种新方法将允许对电池组中所有电池的热失控进行早期检测，使用当前的监测技术是一项非常困难的任务。来自气体传感器的数据可以直接反馈到电池管理系统（BMS）中，以关闭系统或激活预防系统。如果在每个材料分解的“特征”中有足够的分辨率，该方法可能被用来推断由于热循环而导致的细胞降解。现有的BMS在试图检测电池组中的热事件时存在不足，因为它们通常依赖于热电偶等离散温度传感器。在大规模电池组中，由于各种原因，包括成本，制造困难，计算时间等，使用离散传感器监测每个电池是不可行的。相反，一个传感器被用来监测细胞群。然而，这种解决方案是有缺陷的，因为任何热失控事件如果发生在没有直接监测的单元中，则可能无法检测到，直到为时已晚。在包装层进行热失控检测的其他方法包括对高温高压下从电池中排出的HF和其他气体进行气体检测。然而，排气事件发生在热失控过程中，通常为时已晚，无法部署预防策略。预计该方法能够比现有方法更快、更准确地检测包层的热失控。第一年：第一年将包含大部分研究，分为几个部分，解决项目的独特问题。研究将涵盖电池技术、药剂材料、应用技术和气体检测方法。第2-3年：第2-3年主要是实验工作。根据第一年的研究和可用资源，需要进行实验，以确定不同老化状态下各种锂离子电池的表面温度特性。第一学年研究的药剂材料将被制造并用气相色谱仪测试，以确定它们在气体分解和活化温度方面是否符合要求。一旦确定了药剂材料，将进行实验，以找到将其应用于电池的最佳方法。然后将代理材料应用于模拟电池，该电池将通过加热元件加热，以查看其性能是否如预期的那样。这一阶段的气体检测将由气相色谱仪和研究中确定的各种传感器来完成。这将决定哪些传感器适合最终应用。然后将代理材料应用于真实的细胞，这些细胞将有意地置于热失控状态，以评估它们在检测热事件方面的有效性。通过在不同的电池上进行几次不同的循环测试，将研究从涂层中排放的气体的“特征”，以评估确定电池降解的可行性。优化后，该方法将在商用电动汽车电池组上进行测试，以评估该方法在电池组级别的功能。四年级：四年级将主要包括撰写，评估和展示2-3年的实验结果和发现。

The proposed project presents a novel way of monitoring lithium-ion cells inside traction battery packs by using a single gas sensor. Each battery cell in the battery pack will be coated in an agent material which will have a slightly different decomposition signature for each cell. These materials will be designed to "activate" and release gasses at very specific temperatures. By detecting the composition of the released gasses, a single sensor can detect and locate a thermal event occurring at any cell inside the battery pack. Furthermore, the extremely dangerous thermal runaway mechanism that can occur in lithium-ion batteries can be directly characterised by the cell temperature. This new method will allow for early detection of thermal runaway for all cells in a pack, a very difficult task using current monitoring techniques. The data from the gas sensor can be fed directly back into the battery management system (BMS) to shut down the system or activate a prevention system. If there is enough resolution in the 'signature' of each material decomposition, the method could potentially be used to infer the degradation of the cells due to thermal cycling. Existing BMS's fall short when attempting to detect thermal events in a battery pack as they typically rely on discrete temperature sensors such as thermocouples. In large scale battery packs, it is not feasible to monitor each cell with a discrete sensor for various reasons including cost, manufacturing difficulties, computing time etc. Instead, a single sensor is used to monitor groups of cells. This solution is flawed however as any thermal runaway event can go undetected until it is too late if it occurs in a cell which is not directly monitored. Other methods of thermal runaway detection at pack level include gas detection of HF and other gases which are vented from a cell at high temperatures and pressures. The venting event however occurs well into the thermal runaway process and is typically too late to deploy prevention strategies. The proposed method is anticipated to be able to detect thermal runaway at pack level much quicker and more accurately than existing methods.Year 1: The first year will contain mostly research split up into sections which address the distinctive issues of the project. The research will cover battery technology, agent materials, application techniques and gas detection methods. Years 2-3: Years 2-3 will mostly contain experimental work. Depending on the research and available resources in year 1, experiments will need to be carried out to determine the surface temperature characteristics of various lithium-ion batteries under different ageing states. The agent materials researched in year 1 will be manufactured and tested with a gas chromatograph to determine if they function as required in terms of gas decomposition and activation temperature. Once the agent materials are determined, experiments will be carried out to find the best method for applying them to the battery cells. The agent materials will then be applied to dummy batteries which will be heated by a heating element to see if they behave as expected. The gas detection in this stage will be done by both a gas chromatograph and various sensors identified in the research. This will determine which sensors are suitable for the final application. The agent materials will then be applied to real cells which will be purposely put into thermal runaway to assess how effective they are in detecting thermal events. The 'signature' of the gas emitted from the coatings will be investigated through several different cyclic tests on various cells to assess the feasibility of determining the cell degradation. Once optimised, the method will be tested on a commercial EV battery pack to assess the functionality of the method at pack level. Year 4: Year 4 will mostly contain writing up, evaluating, and presenting the results and findings from the experiments in years 2-3.