keilongw / battery-state-estimation Goto Github PK

Can you help me build it and I can pay the copyright fee?

Hello trying to run your code getting an error could you please help .

Hello, I encountered an issue while training the model while trying to replicate your project.NotImplementedError: Layer ModuleWrapper was created by passingnon-serializable argument values in __init__(),and therefore the layer must override get_config() inorder to be serializable. Please implement get_config().Can you provide an answer

First, thank you for all your work in putting out this excellent repository! This has been quite insightful.

I'm reviewing the lg_dataset.py data processing module. I understand some of the functions, but I'm not clear on the others.

1. _get_data - computes SOC, assigns Voltage, Current, Temperature to 'x', SOC labels to 'y'
2. _time_string_to_seconds - converts the 'Program Time' column to seconds
3. get_discharge_whole_cycle - creates the test and train datasets using the '_get_data' function
4. get_stateful_cycle - not sure what this does
5. _to_padded_cycle - not sure what this does
6. _split_cycle - not sure what this does
7. _split_to_multiple_step - not sure what this does
8. keep_only_y_end - not sure what this does

Is my understanding of these functions correct and could you help describe the ones I'm unsure about?

Overall, it looks like you can process the TRAIN and TEST datasets in two ways - as (1) 'get discharge multiple step' or as (2) 'stateful cycle'. What does stateful cycle mean, and are there scenarios where you use one over the other?

Thanks.

Hello @KeiLongW,

I'm going thru the result models you've put up (thanks for those!). I notice that for the experiments - lstm_soc_percentage_lg_result and lstm_soc_capacity_lg_result.

The predictions on the training dataset are accurate, however the SOC predictions on the test dataset differ widely from actual SOC. I notice this when I run training on the corresponding experiments as well.

Any ideas why that could be? I've attached screenshots of the predictions I'm seeing.

Hello, when reforking according to your program, why does the following error appear "IndexError: boolean index did not match indexed array along dimension 0; dimension is 84737 but corresponding boolean dimension is 84734"? Thank you so much

D:\Anaconda\envs\DL\python.exe C:/Users/24451/Desktop/battery-state-estimation/battery-state-estimation/data_processing/lg_dataset.py
Traceback (most recent call last):
File "C:\Users\24451\Desktop\battery-state-estimation\battery-state-estimation\data_processing\lg_dataset.py", line 253, in
cycles = lg_data.get_discharge_whole_cycle(train_names, test_names)
File "C:\Users\24451\Desktop\battery-state-estimation\battery-state-estimation\data_processing\lg_dataset.py", line 16, in get_discharge_whole_cycle
train = self._get_data(train_names, output_capacity, output_time)
File "C:\Users\24451\Desktop\battery-state-estimation\battery-state-estimation\data_processing\lg_dataset.py", line 24, in _get_data
cycle = pd.read_csv(self.path + name + '.csv', skiprows=30)
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\parsers.py", line 610, in read_csv
return _read(filepath_or_buffer, kwds)
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\parsers.py", line 462, in _read
parser = TextFileReader(filepath_or_buffer, **kwds)
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\parsers.py", line 819, in init
self._engine = self._make_engine(self.engine)
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\parsers.py", line 1050, in _make_engine
return mapping[engine](self.f, **self.options) # type: ignore[call-arg]
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\parsers.py", line 1867, in init
self._open_handles(src, kwds)
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\parsers.py", line 1362, in _open_handles
self.handles = get_handle(
File "D:\Anaconda\envs\DL\lib\site-packages\pandas\io\common.py", line 642, in get_handle
handle = open(
FileNotFoundError: [Errno 2] No such file or directory: './data/LG 18650HG2 Li-ion Battery Data/LG_HG2_Original_Dataset_McMasterUniversity_Jan_2020/25degC/551_LA92.csv'

第一步和第二部我的环境都没有问题，到第三步我应该怎么操作才可以运行起来？
求答复，谢谢

Hello,I used your code unibo_powertools_data.py.I find that PROCESSED_RESULT_DATA_PATH = 'data/unibo-powertools-dataset/test_result_final.csv',the file'test_result_final.csv' is already included in unibo dataset,if I run this code,origin data will be replaced.Besides,'test_result.csv' is too big to run in all the unibo.ipynb(the process is died).What can I do to solve problems above?

Hello, I used your code to run an experiment, but I encountered overfitting. The trained network performs well on the training set but very poorly on the test set. Have you ever encountered this situation? I ran the experiment in experiments/lg/lstm_soc_percentage_lg.ipynb.