-
Notifications
You must be signed in to change notification settings - Fork 190
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Example] Add battery-electrochemical-performance prediction model #967
base: develop
Are you sure you want to change the base?
Changes from all commits
4c8a961
21d7ffc
6e742a7
61ba5c2
f76b48a
19d6858
8a1a349
2762835
0fa9a14
6cf2951
6d7c2ef
b45fe5f
152956f
eabdc43
b5d7636
473b373
5336d15
d3fa030
ba9fcaf
b3ae407
77a6f6b
627e4b6
f5fa76c
8efb7ca
d3aba2c
f106ca8
98d3e72
5e56111
04d2d89
183c888
7cabf19
8d756ce
583f2fb
6823a3e
7fe6d3c
08ec87a
c713103
eed03c0
1ff40c2
73d725f
ce4b352
7c7e4fc
16c86c6
145445a
e13c8f1
3baffd2
45b1474
bb1d3d1
e46008a
c489261
10abaa6
7736f44
aa5eaa1
1b96d9b
ff7d715
4dd897a
03cab2f
77eeadb
effc838
a52ce7c
47356b3
88b63d3
8028ad6
6b2cc63
c752895
fc8f131
18ea97a
434dfc4
b280169
f64029d
cd5345b
6e82a63
2c2bc8d
10015f7
b4579ad
cda91fd
483a3fa
e372903
754cbaa
9ccf46c
54f8b8d
17eff79
9b62a10
f0eaa6a
d7e4991
7827f42
8d4369a
2af9ae8
050c30d
fff722e
89aeb4d
a81d083
5862588
9261c69
f048a4f
fcfc38e
6503adb
8ad8d0c
53d9a20
05bb35c
cb21ea4
0a1c4a0
77e738e
277e41e
95e979e
268aae9
344d60f
46e204a
fa00f6d
0b432a8
b94c9b1
30da823
c5118d9
8ddc2f4
0205128
1c93d88
eba289d
53a3a2a
b96fc7b
a331946
97ca920
5cd0162
7f66831
0cf728f
6d2e17c
f19a5c6
3f3d2f9
c61a5f6
fae0f16
23dd4e5
6d375c1
cfae240
47c433b
f0d3ae9
e0b7d15
fe175c4
8f1f27d
60a6369
7c04bf8
369b726
61be241
6a96beb
f1a0336
c29678e
98bfb30
6b81879
4d44ed5
25725c6
f874bf2
d6047cf
3d70cbe
a1830a1
c57a97a
fceb3f2
bfee4f5
f5dac19
62608a8
c46a933
33b6597
e5cdf29
452ba2f
ce050a3
99d5170
3fb6449
30a2736
1cd67d3
a880f4b
9b91ca7
085b458
ec28348
1e61422
0d4e2ea
9e9d147
3113eb5
4171173
ee53eeb
1162cc6
c8c83b4
fba7481
76e16ce
aea71b4
e96e93d
2edfaa5
41ee16e
25ff28a
6dd6a98
a2240ac
985ed10
c871670
1deb3bc
8997f72
7f8817d
c2134a6
d62cffd
02db125
0d61c06
9bb8dc0
6ba4bee
af9dab1
854dc9e
4735514
fc5bd69
62d0457
fbad74e
5f3a1a1
e3accd4
fcbfdfa
2809f99
5840712
8641faa
2834c8a
eaa3982
d67a5b9
957ac7e
93117ae
1cf8019
40dc557
4c8a473
4154f32
add2e32
dba9c3a
46cc652
95f0f1b
4d6a057
e98b3cf
a2f0da3
22a6bb7
7380a0f
170ab7e
9ae0d7e
016c31c
6aa419c
52eecac
7178d01
30c9aa4
e16c0ae
221f1c1
3cb2ae1
7dc8268
854707d
fb28d61
d091f24
c44cf29
83fc2c1
065b0f5
c8dc22c
ea52116
5b3b911
0791471
e90e1eb
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,90 @@ | ||
# Battery_LI(锂离子电池电极材料性能预测) | ||
|
||
## 背景简介 | ||
|
||
锂离子电池(Lithium-ion Battery, LIB)作为现代储能技术的核心,广泛应用于消费电子、电动汽车、以及可再生能源的存储等领域。电极材料是锂离子电池性能的关键,其性能直接决定了电池的能量密度、功率密度、寿命、和安全性。然而,电极材料的研发是一个复杂且耗时的过程,通常需要实验测试和理论计算相结合,这对时间和资源的消耗非常大。 | ||
|
||
## 模型原理 | ||
|
||
该多层感知器(MLP)模型旨在利用从材料项目(Materials Project)数据集中提取的特征,预测锂离子电池电极材料的电化学性能。输入特征包括化学计量属性、晶体结构特性、电子结构属性和其他电池属性。输出为平均电压、比能量和比容量。 | ||
|
||
## 数据集介绍 | ||
关于数据集,请查看该文件,MP_data_down_loading(train+validate).csv | ||
|
||
数据读取需要额外安装依赖 `bayesian-optimization`,请运行安装命令 `pip install bayesian-optimization`。 | ||
|
||
## 模型 | ||
|
||
要查看该模型的具体实现,请参考以下代码文件:`MLP_LI.py` | ||
|
||
## 模型训练命令 | ||
=== "模型训练命令" | ||
|
||
``` sh | ||
python MLP_LI.py | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. |
||
``` | ||
|
||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 请添加模型评估命令,即直接基于下面训练好的模型进行评估 |
||
## 训练好的模型权重文件 | ||
|
||
| 预训练模型 | | ||
|-----------------------------------| | ||
| [MLP_LI_pretrained.pdparams](https://paddle-org.bj.bcebos.com/paddlescience/models/MLP_LI/MLP_LI_pretrained.pdparams) | | ||
|
||
## 完整代码 | ||
|
||
``` py linenums="1" title="examples/MLP_LI/MLP_LI.py" | ||
--8<-- | ||
examples/MLP_LI/MLP_LI.py | ||
--8<-- | ||
``` | ||
|
||
## 模型性能 | ||
|
||
模型在测试集上的表现如下: | ||
|
||
- **Test Loss**: 0.0058 | ||
|
||
- **VRMSE 电压**: 0.73 | ||
- **CRMSE 比容量**:165.01 | ||
- **ERMSE 比能量**: 238.64 | ||
- **Average RMSE 平均值**:134.79 | ||
|
||
此外,模型在各个输出指标上的平均绝对误差(MAE)如下: | ||
|
||
- **VMAE 电压**: 0.55 | ||
- **CMAE 比容量**: 73.34 | ||
- **EMAE 比能量**: 180.10 | ||
- **Average MAE 平均值**: 84.66 | ||
|
||
这些结果表明模型在预测电压方面具有较高的精度,而在预测比容量和比能量方面还有一定的改进空间。 | ||
|
||
### 图表 | ||
|
||
#### 1. 电压的性能预测(原始尺度) | ||
此图展示了电压的性能预测。预测值与真实值的比较用于评估模型的准确性。 | ||
|
||
 | ||
|
||
#### 2. 性能预测(原始尺度) | ||
此图展示了模型对所有三个电化学性能(电压、比能量和比容量)的整体预测表现。 | ||
|
||
 | ||
|
||
#### 3. 初始训练损失 | ||
以下图显示了在初始训练阶段的训练和验证损失变化情况(按Epochs)。 | ||
|
||
 | ||
|
||
## 结论 | ||
该 MLP 模型在提供的数据集上表现出较强的预测能力,尤其是在电压的预测上。然而,在比容量和比能量的预测上还有进一步改进的空间。未来可以通过更丰富的特征工程、更复杂的模型架构以及优化的超参数调整来提高模型的预测性能。 | ||
|
||
## 下一步 | ||
1. 考虑增加额外的特征或进行特征工程,以提高模型预测的准确性。 | ||
2. 尝试不同的神经网络架构或优化策略,以改进性能。 | ||
3. 继续进行超参数优化,以获得更好的模型性能。 | ||
|
||
|
||
## 参考资料 | ||
|
||
Yang, X., Li, Y., Liu, Z., & Zhang, W. (2022) | ||
(https://doi.org/10.1016/j.gee.2022.10.002) |
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 该权重文件已上传:https://paddle-org.bj.bcebos.com/paddlescience%2Fmodels%2FMLP_LI_pretrained.pdparams 可以在PR中删除这个文件,然后相关的权重链接更新为该url即可 |
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. |
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. |
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件能使用vscode的markdown格式化插件格式化一下吗?如果没有安装vscode的话就算了