WO2024060539A1 - 动力电池的年度碳排放量估算方法及装置 - Google Patents
动力电池的年度碳排放量估算方法及装置 Download PDFInfo
- Publication number
- WO2024060539A1 WO2024060539A1 PCT/CN2023/081880 CN2023081880W WO2024060539A1 WO 2024060539 A1 WO2024060539 A1 WO 2024060539A1 CN 2023081880 W CN2023081880 W CN 2023081880W WO 2024060539 A1 WO2024060539 A1 WO 2024060539A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- power battery
- annual
- weight
- production volume
- carbon emissions
- Prior art date
Links
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 136
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 135
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000004519 manufacturing process Methods 0.000 claims abstract description 181
- 238000013145 classification model Methods 0.000 claims abstract description 37
- 239000000463 material Substances 0.000 claims description 92
- 238000012549 training Methods 0.000 claims description 51
- 238000003062 neural network model Methods 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 21
- 230000015654 memory Effects 0.000 claims description 21
- 239000000470 constituent Substances 0.000 claims description 19
- 238000003860 storage Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 239000007773 negative electrode material Substances 0.000 claims description 6
- 239000007774 positive electrode material Substances 0.000 claims description 6
- 238000012731 temporal analysis Methods 0.000 claims description 6
- 238000000700 time series analysis Methods 0.000 claims description 6
- 239000003792 electrolyte Substances 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 239000000306 component Substances 0.000 description 23
- 238000004364 calculation method Methods 0.000 description 9
- 230000000306 recurrent effect Effects 0.000 description 9
- 238000012360 testing method Methods 0.000 description 8
- 230000006403 short-term memory Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- -1 lithium hexafluorophosphate Chemical compound 0.000 description 3
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 2
- SECXISVLQFMRJM-UHFFFAOYSA-N N-Methylpyrrolidone Chemical compound CN1CCCC1=O SECXISVLQFMRJM-UHFFFAOYSA-N 0.000 description 2
- 239000002033 PVDF binder Substances 0.000 description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 229920002981 polyvinylidene fluoride Polymers 0.000 description 2
- KMTRUDSVKNLOMY-UHFFFAOYSA-N Ethylene carbonate Chemical compound O=C1OCCO1 KMTRUDSVKNLOMY-UHFFFAOYSA-N 0.000 description 1
- 101001121408 Homo sapiens L-amino-acid oxidase Proteins 0.000 description 1
- 102100026388 L-amino-acid oxidase Human genes 0.000 description 1
- 239000004698 Polyethylene Substances 0.000 description 1
- 239000004743 Polypropylene Substances 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 239000011889 copper foil Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- IEJIGPNLZYLLBP-UHFFFAOYSA-N dimethyl carbonate Chemical compound COC(=O)OC IEJIGPNLZYLLBP-UHFFFAOYSA-N 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000011888 foil Substances 0.000 description 1
- 239000010439 graphite Substances 0.000 description 1
- 229910002804 graphite Inorganic materials 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- GELKBWJHTRAYNV-UHFFFAOYSA-K lithium iron phosphate Chemical compound [Li+].[Fe+2].[O-]P([O-])([O-])=O GELKBWJHTRAYNV-UHFFFAOYSA-K 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229920000573 polyethylene Polymers 0.000 description 1
- 229920001155 polypropylene Polymers 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
Definitions
- This application relates to the technical field of carbon emission estimation, for example, to a method and device for estimating annual carbon emissions of a power battery.
- New energy vehicles have zero emissions, but carbon emissions are generated during the production of raw materials and batteries. As the new energy vehicle market rapidly expands, the environmental protection and carbon emissions issues of power batteries, its core components, are attracting more and more attention.
- This application provides a method and device for estimating annual carbon emissions of power batteries to improve the efficiency and accuracy of carbon emission estimation.
- This application provides a method for estimating annual carbon emissions of power batteries, including: obtaining annual sales data of new energy vehicles, and inputting the annual sales data into a pre-trained annual production forecast model of power batteries, so that the The pre-trained annual power battery production forecast model outputs the annual power battery forecast production volume; the annual power battery forecast production volume is input into the pre-trained power battery weight classification model, so that the pre-trained power battery weight classification model Based on the difference in the total weight of a single power battery, output the predicted production volume of the first weight power battery corresponding to multiple total weights; obtain the first carbon emissions corresponding to the predicted production volume of the first weight power battery, and integrate all first carbon emissions amount to obtain the annual carbon emissions.
- obtaining the first carbon emissions corresponding to the predicted production volume of the first weight power battery includes: obtaining all the constituent materials of a single power battery in the predicted production volume of the first weight power battery, And obtain the first material weight corresponding to each component material; according to the first material weight and the predicted production volume of the first weight power battery, obtain each component material in the predicted production volume of the first weight power battery.
- the corresponding second material weight obtain the standard unit carbon emissions corresponding to each component material, and obtain the predicted production volume of the first weight power battery based on the second material weight and the standard unit carbon emissions.
- the first material carbon emissions corresponding to each component material are integrated to obtain the first carbon emissions corresponding to the predicted production volume of the first weight power battery.
- the pre-training process of the annual power battery production forecast model includes: obtaining historical annual sales data of new energy vehicles and historical annual production volume of power batteries corresponding to different years, and obtaining historical samples based on time series analysis. Data set; perform random sampling processing on the historical sample data set to obtain a training set; input the training set into a neural network for forward propagation and backward propagation to obtain a power battery annual production prediction model.
- the pre-training process of the power battery weight classification model includes: obtaining the historical annual production volume of power batteries corresponding to different years, and based on the difference in the total weight of a single power battery, calculating the historical annual production volume of the power battery. Classify the quantity to obtain the historical production volume of the first-weight power battery corresponding to various total weights; use the historical annual production volume of the power battery as the input of the neural network model, and classify the first-weight power battery corresponding to the various total weights The historical production volume is used as the output of the neural network model until the training of the neural network model converges, and a power battery weight classification model is obtained.
- all constituent materials of the single power battery include positive electrode materials, negative electrode materials, electrolytes, separators, casings and battery management systems BMS.
- This application also provides an annual carbon emission estimation device for power batteries, including: an annual power battery forecast production acquisition module, a power battery weight classification module and an annual carbon emission acquisition module; wherein the annual power battery forecast production
- the volume acquisition module is configured to obtain the annual sales data of new energy vehicles, and input the annual sales data into the pre-trained power battery annual production forecast model, so that the pre-trained power battery annual production forecast model outputs annual power Battery predicted production volume
- the power battery weight classification module is configured to input the annual power battery predicted production volume into a pre-trained power battery weight classification model, so that the pre-trained power battery weight classification model is based on a single Depending on the total weight of the power battery, output the predicted production volume of the first weight power battery corresponding to various total weights;
- the annual carbon emission acquisition module is configured to obtain the first carbon emission corresponding to the predicted production volume of the first weight power battery, integrate all the first carbon emissions, and obtain the annual carbon emission.
- the annual carbon emission acquisition module is configured to obtain the The first carbon emission amount corresponding to the predicted production volume of the first weight power battery includes: obtaining all the constituent materials of a single power battery in the predicted production volume of the first weight power battery, and obtaining the first material weight corresponding to each constituent material. ;According to the weight of the first material and the predicted production volume of the first weight power battery, obtain the second material weight corresponding to each component material in the predicted production volume of the first weight power battery; Obtain the weight of each component The standard unit carbon emissions corresponding to the constituent materials. Based on the second material weight and the standard unit carbon emissions, the first material carbon emissions corresponding to each constituent material in the predicted production volume of the first weight power battery are obtained. , integrating all the first material carbon emissions to obtain the first carbon emissions corresponding to the predicted production volume of the first weight power battery.
- the pre-training process of the annual power battery production forecast model in the annual power battery forecast production acquisition module includes: acquiring the historical annual sales data of new energy vehicles and the historical annual power battery data corresponding to different years. production volume to obtain a historical sample data set based on time series analysis; perform random sampling processing on the historical sample data set to obtain a training set; input the training set into the neural network for forward propagation and backward propagation to obtain power Annual battery production forecast model.
- the pre-training process of the power battery weight classification model in the power battery weight classification module includes: the historical annual production volume of power batteries corresponding to different years, based on the difference in the total weight of a single power battery, The historical annual production volume of the power battery is classified to obtain the historical production volume of the first weight power battery corresponding to various total weights; the historical annual production volume of the power battery is used as the input of the neural network model, and the various total weight corresponding The historical production volume of the first weight power battery is used as the output of the neural network model until the training of the neural network model converges, and a power battery weight classification model is obtained.
- all constituent materials of the single power battery include positive electrode materials, negative electrode materials, electrolytes, separators, casings and BMS.
- This application also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor.
- a terminal device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor.
- the processor executes the computer program, any of the above is implemented.
- the computer-readable storage medium includes a stored computer program.
- the device where the computer-readable storage medium is located is controlled to execute any one of the above steps.
- FIG1 is a flow chart of an embodiment of a method for estimating annual carbon emissions of a power battery provided in the present application
- Figure 2 is a schematic structural diagram of an embodiment of an annual carbon emission estimation device for a power battery provided by this application;
- Figure 3 is a schematic structural diagram of a terminal device provided by this application.
- Figure 1 is a schematic flow chart of an embodiment of a method for estimating annual carbon emissions of a power battery provided by this application. As shown in Figure 1, the method includes steps 101 to 103.
- Step 101 Acquire annual sales data of new energy vehicles, and input the annual sales data into a pre-trained power battery annual production forecasting model, so that the pre-trained power battery annual production forecasting model outputs an annual power battery forecasting production volume.
- the pre-training process obtains historical sample data sets based on time series analysis by obtaining historical annual sales data of new energy vehicles and historical annual production volume of power batteries corresponding to different years. That is, the obtained historical annual sales data of new energy vehicles and the historical annual production volume of power batteries are arranged in the order of years to obtain the data sequence and form a historical sample data set.
- a time label is set for each historical sample data in the historical sample data set according to the year information of data acquisition, wherein the time label is the year data.
- the historical sample data set is randomly sampled to obtain a training set; the historical sample data set is divided into a training set and a test set according to a preset ratio, and based on the number of historical sample data in the divided training set, from A corresponding number of historical sample data are randomly extracted from the historical sample data set to form a training set, and the remaining part is used as a test set; where the preset ratio is 8:2.
- the training set is input into a neural network model for forward propagation and backward propagation to obtain a power battery annual production prediction model.
- the training set is input into a long-short-term memory recurrent neural network model for training to obtain the original power battery annual production prediction model.
- the long-short-term memory recurrent neural network model is a time-recursive neural network model suitable for processing and forecasting important events that have time series with relatively long time series intervals and delays.
- the long short-term memory recurrent neural network model includes a three-layer network structure of an input layer, a hidden layer and an output layer.
- the input layer is the first layer of the long-short-term memory recurrent neural network model, which is used to receive external signals, that is, it is responsible for receiving the historical annual sales data of new energy vehicles and the historical annual production volume of power batteries in the training set;
- the output layer is the long-short-term memory recurrent neural network.
- the last layer of the network model is used to output signals to the outside world, that is, it is responsible for outputting the calculation prediction results of the long short-term memory recurrent neural network model;
- the hidden layer is the multi-layer layer in the long short-term memory recurrent neural network model other than the input layer and the output layer.
- a layer is used to process the historical annual sales data of new energy vehicles and the historical annual production volume of power batteries in the training set, and obtain the calculation results of the long and short-term memory recurrent neural network model.
- using the long short-term memory recurrent neural network model for model training increases the temporal sequence of the training data, thereby improving the accuracy of the prediction model.
- the training set is obtained and trained based on the forward propagation algorithm to obtain the first state parameter of the original power battery annual production prediction model, where the first state parameter refers to the initial iteration process during model training based on the training data.
- the error calculation of the first state parameter is obtained and based on the backward propagation algorithm to obtain the second state parameter of the original power battery annual production prediction model. Based on the second state parameter, the original power battery annual production forecast model is obtained. Production forecasting model.
- the backpropagation algorithm is In the formula, O t represents the time t and corresponding true value.
- the cross-entropy loss function is used to calculate the partial derivative of each layer, and the three weight parameters U, V and W are updated based on the partial derivative to obtain the adjusted second state parameter; wherein the partial derivative includes
- the original power battery annual production prediction model is tested based on the test set to obtain the power battery annual production prediction model.
- the historical annual sales data of new energy vehicles in any year in the test set are sequentially input into the original power battery annual production forecast model, so that the original power battery annual production forecast model sequentially outputs the power battery annual forecast production volume, and the power battery annual production forecast model is The annual predicted production volume is compared with the historical annual production volume of the power battery corresponding to the year to obtain the prediction error value. It is judged whether the prediction error value is within the preset error range. If so, the original power battery annual production volume is calculated. The prediction model is used as the power battery annual production prediction model; if not, the original power battery annual production prediction model will continue to be trained until the prediction error value is within the prediction error range.
- Step 102 Input the annual power battery forecast production volume into the pre-trained power battery weight score.
- the pre-trained power battery weight classification model outputs the predicted production volume of the first weight power batteries corresponding to multiple total weights based on the different total weights of a single power battery.
- the weight of the power battery is affected by the rated voltage and capacity of the power battery. Different rated voltages and capacities will result in different weights of the power battery. Therefore, based on the different new energy vehicles, the corresponding weight of the power battery is also different. There may be differences. Based on this, it is also necessary to pre-build a power battery weight classification model so that the predicted annual power battery production volume predicted based on the power battery weight classification model is classified according to the total weight of a single power battery.
- the training process of the power battery weight classification model is to obtain the historical annual production volume of the power battery corresponding to the same year, and classify the historical annual production volume of the power battery based on the difference in the total weight of a single power battery, and obtain multiple The historical production volume of the first-weight power battery corresponding to the various total weights; the historical annual production volume of the power battery is used as the input of the neural network model, and the historical production volume of the first-weight power battery corresponding to the various total weights is used as the neural network model The output of the model is used until the training of the neural network model converges, and a power battery weight classification model is obtained.
- a historical production volume data set is generated, and the historical production volume data set is divided into a first training set and a first test set according to a preset ratio; wherein the prediction ratio is 8:2.
- the neural network model is trained based on the first training set, and the original power battery weight classification model is determined by recording the loss value and accuracy value during the model training process.
- model training is intuitive and controllable.
- the loss value and accuracy value obtained at each step of training are recorded.
- the test set is used to calculate the model accuracy every 5 steps to obtain the verification loss value and Accuracy value, and record the verification loss value and accuracy value in TensorBoard, so that the loss value and accuracy value can be drawn into a change curve chart to measure the model recognition effect under the current training progress.
- the loss change threshold and the maximum number of iterations are also set.
- the loss change threshold and the maximum number of iterations are used to determine whether to stop training the model; when the model training does not reach the maximum number of iterations, and the change in the loss value is within a certain
- the loss change threshold has not been reached during the period, it is judged whether the verification accuracy value begins to decline. If not, continue to train the model. If so, it means that the fitting degree of the model has reached the optimal peak value at this time. Next If you train again, the accuracy of the verification data set will continue to decrease, leading to model overfitting, so training should be stopped at this time.
- the original power battery weight classification model is tested based on the first test set to obtain the power battery weight classification model.
- the historical annual production volume of power batteries in any year in the first test set is sequentially input into the original power battery weight classification model, so that the original power battery weight classification model outputs multiple total weights in sequence
- For the corresponding historical predicted production volume of the first weight power battery compare the historical predicted production volume of the first weight power battery with the historical production volume of the first weight power battery corresponding to the year to obtain the first prediction error value, and determine the Whether the first prediction error value is within the first preset error range, if so, the original power battery weight classification model will be used as the power battery weight classification model; if not, then the original power battery weight classification model will continue to be used. Training is performed until the first prediction error value is within the first prediction error range.
- Step 103 Obtain the first carbon emissions corresponding to the preset production volume of the first weight power battery, integrate all the first carbon emissions, and obtain the annual carbon emissions.
- all constituent materials of a single power battery in the predicted production volume of the first weight power battery are obtained, and the first material weight corresponding to each constituent material is obtained.
- all constituent materials of the single power battery include positive electrode materials, negative electrode materials, electrolytes, separators, casings and battery management systems (Battery Management System, BMS).
- BMS Battery Management System
- the positive electrode material includes lithium iron phosphate, polyvinylidene fluoride, N-methylpyrrolidone and aluminum foil;
- the negative electrode material includes graphite, polyvinylidene fluoride, N-methylpyrrolidone, and copper foil;
- the electrolytic The liquid includes lithium hexafluorophosphate, ethylene carbonate and dimethyl carbonate;
- the separator includes polypropylene and polyethylene;
- the housing includes aluminum sheets; and the BMS includes circuit boards, steel sheets and copper sheets.
- the first material weight corresponding to each component material in a single power battery is obtained by obtaining the material mass required by the standard unit corresponding to each component material, based on the total weight of the single power battery and the required standard unit. Material mass, the first material weight corresponding to each component material in a single power battery is obtained.
- the second material weight corresponding to each component material in the predicted production volume of the power battery with the first weight is obtained.
- the weight of the first material is the weight of a single required component material in a single power battery
- the standard unit carbon emissions corresponding to each component material is obtained, and based on the weight of the second material and the standard unit carbon emissions, the predicted production volume of the first weight power battery is obtained.
- the first material carbon emissions corresponding to each component material are integrated to obtain the first carbon emissions corresponding to the predicted production volume of the first weight power battery.
- E i M i ⁇ e i ; wherein E i is the carbon emission of the first material corresponding to the i-th required component material, in kg, and e i is the standard unit carbon emission corresponding to the i-th required component material.
- the calculated carbon emissions of the first material corresponding to all required constituent materials are superimposed to form the first carbon emissions corresponding to the predicted production volume of the first weight power battery.
- the calculation process is as follows: Among them, wt is the first carbon emission corresponding to the predicted production volume of the first weight power battery of the tth group.
- all the first carbon emissions are integrated to obtain the annual carbon emissions; all the first carbon emissions are superimposed to form the annual carbon emissions of the annual power battery predicted production volume, so as to realize the annual carbon emissions.
- Forecast of annual carbon emissions; the calculation process of annual carbon emissions is as follows: Among them, W is the annual carbon emissions in kg.
- the annual carbon emissions corresponding to the future target year can be compared with the annual carbon emissions corresponding to the current year to obtain the annual carbon emissions growth trend.
- the annual carbon emission growth trend can be upward or downward, so that relevant enterprises can timely understand carbon emission-related information and provide important green support for the enterprise's subsequent carbon emission treatment.
- Figure 2 is a schematic structural diagram of an embodiment of an annual carbon emission estimation device for power batteries provided by this application. As shown in Figure 2, the device includes an annual power battery forecast production volume acquisition module 201, Power battery weight classification module 202 and annual carbon emissions acquisition module 203.
- the annual power battery predicted production volume acquisition module 201 is configured to obtain annual sales data of new energy vehicles, and input the annual sales data into the pre-trained power battery annual production forecast model, so that the pre-trained power battery
- the annual battery production forecast model outputs the annual power battery forecast production volume
- the power battery weight classification module 202 is configured to input the annual power battery forecast production volume into the pre-trained power battery weight classification model, so that the pre-trained power battery weight classification model
- the trained power battery weight classification model outputs the predicted production volume of the first weight power battery corresponding to multiple total weights based on the total weight of a single power battery
- the annual carbon emissions acquisition module 203 is configured to obtain the first weight
- the first carbon emissions corresponding to the production volume of the power battery are predicted, and all the first carbon emissions are integrated to obtain the annual carbon emissions.
- the annual carbon emissions acquisition module 203 is configured to obtain the first carbon emissions corresponding to the predicted production volume of the first weight power battery, including: obtaining the predicted production volume of the first weight power battery. All constituent materials of a single power battery, and obtain the first material corresponding to each constituent material Material weight; according to the first material weight and the first weight power battery predicted production volume, obtain the second material weight corresponding to each component material in the first weight power battery predicted production volume; obtain each component material Corresponding standard unit carbon emissions, based on the second material weight and the standard unit carbon emissions, obtain the first material carbon emissions corresponding to each component material in the predicted production volume of the first weight power battery, and integrate For all the carbon emissions of the first material, the first carbon emissions corresponding to the predicted production volume of the first weight power battery are obtained.
- the pre-training process of the power battery annual production forecast model in the annual power battery forecast production acquisition module 201 includes: obtaining historical annual sales data of new energy vehicles and historical annual production of power batteries corresponding to different years to obtain a historical sample data set based on time series analysis; performing random sampling on the historical sample data set to obtain a training set; and inputting the training set into a neural network for forward propagation and backward propagation to obtain a power battery annual production forecast model.
- the pre-training process of the power battery weight classification model in the power battery weight classification module 202 includes: obtaining the historical annual production volume of power batteries corresponding to different years, and based on the difference in the total weight of a single power battery, The historical annual production volume of power batteries is classified to obtain the historical production volume of the first weight power battery corresponding to various total weights; the historical annual production volume of power batteries is used as the input of the neural network model, and the historical production volume corresponding to the various total weights is obtained The historical production volume of the first weight power battery is used as the output of the neural network model until the training of the neural network model converges, and a power battery weight classification model is obtained.
- all constituent materials of the single power battery include positive electrode materials, negative electrode materials, electrolytes, separators, casings and BMS.
- the above embodiments of the annual carbon emission estimation device for power batteries are only illustrative, in which the modules described as separate components may or may not be physically separated, and the components shown as modules may It may or may not be a physical unit, i.e. it may be located in one place, or it may be distributed over multiple network units. Some or all of the modules can be selected according to actual needs to implement the solution of this embodiment.
- an annual carbon emission estimation terminal device of the power battery includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor.
- the processor executes the computer program, it implements the power battery of any embodiment of the present application.
- the computer program may be divided into one or more modules,
- the one or more modules are stored in the memory and executed by the processor to complete the application.
- the one or more modules may be a series of computer program instruction segments capable of completing specific functions.
- the instruction segments are used to describe the execution process of the computer program in the annual carbon emission estimation terminal device of the power battery.
- the terminal device for estimating the annual carbon emissions of the power battery may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc.
- the terminal device for estimating the annual carbon emissions of the power battery may include, but is not limited to, a processor and a memory.
- the so-called processor can be a central processing unit (Central Processing Unit, CPU), or other general-purpose processor, digital signal processor (Digital Signal Processor, DSP), application specific integrated circuit (Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
- the general processor can be a microprocessor or the processor can be any conventional processor, etc.
- the processor is the control center of the annual carbon emission estimation terminal device of the power battery, using a variety of interfaces and line connections. The annual carbon emissions of the entire power battery are estimated for multiple parts of the terminal equipment.
- the memory may be configured to store the computer program and/or module, and the processor implements the process by running or executing the computer program and/or module stored in the memory, and calling data stored in the memory.
- the annual carbon emissions of power batteries are estimated for various functions of end equipment.
- the memory may mainly include a stored program area and a stored data area, wherein the stored program area may store an operating system, at least one application required for a function, etc.; the stored data area may store data created based on the use of the mobile phone, etc.
- the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card , Flash Card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
- non-volatile memory such as hard disk, memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card , Flash Card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
- another embodiment of the present application provides a storage medium, the storage medium includes a stored computer program, wherein when the computer program is running , controlling the device where the storage medium is located to execute the annual carbon emission estimation method of the power battery according to any embodiment of this application.
- the above-mentioned storage medium is a computer-readable storage medium
- the computer program includes computer program code.
- the computer program code may be in the form of source code, object code, executable file, or some intermediate form.
- the computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (Read-Only Memory, ROM) , random access memory (Random Access Memory, RAM), electrical carrier signals, telecommunications signals and software distribution media wait.
- the content contained in the computer-readable medium can be appropriately added or deleted according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not Including electrical carrier signals and telecommunications signals.
- this application provides a method and device for estimating annual carbon emissions of power batteries.
- the obtained annual sales data of new energy vehicles are input into the pre-trained annual production forecast model of power batteries, so that the pre-trained power
- the annual battery production forecast model outputs the annual power battery forecast production volume;
- the annual power battery forecast production volume is input into the pre-trained power battery weight classification model, so that the pre-trained power battery weight classification model is based on the difference in the total weight of a single power battery , output the predicted production volume of the first weight power battery corresponding to various total weights; obtain the first carbon emissions corresponding to the predicted production volume of the first weight power battery, integrate all the first carbon emissions, and obtain the annual carbon emissions.
- This application helps companies predict the production volume of power batteries based on the annual sales volume of new energy vehicles, and classifies the predicted production volume of power batteries based on different weights of power batteries, which can improve the efficiency and accuracy of carbon emission estimates.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Manufacturing & Machinery (AREA)
- Secondary Cells (AREA)
Abstract
一种动力电池的年度碳排放量估算方法及装置,通过将获取的新能源汽车的年度销售数据输入到预训练的动力电池年度生产预测模型中,以使预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;将年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;获取第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
Description
本申请要求在2022年09月23日提交中国专利局、申请号为202211177780.8的中国专利申请的优先权,该申请的全部内容通过引用结合在本申请中。
本申请涉及碳排放量估算的技术领域,例如是涉及一种动力电池的年度碳排放量估算方法及装置。
新能源汽车是零排放,但原材料及电池的生产过程中会产生碳排;随着新能源汽车市场规模迅速扩大,其核心部件动力电池的环保和碳排放问题也越来越受关注。
为面对国家提出的“碳中和”目标,不少企业在开始关注动力电池在生产过程中的碳排放,以为后续进行碳排放量处理提供数据支撑,但目前,对于动力电池在生产过程中的碳排放,通过在生产车间装设碳排放量检测器,用以获取动力电池在生产阶段中产生的碳排放量,当该操作无法实现对动力电池生产阶段的碳排放量进行提前预估,且仅能在生产完成后,才能输出检测到的碳排放量,所需时间长,效率低下;同时,现有中还存在基于获取动力电池的生产量直接进行碳排放量计算,忽视了不同动力电池之间重量不同,导致的碳排放量不同的问题,使得计算出来的碳排放量误差较大。
发明内容
本申请提供一种动力电池的年度碳排放量估算方法及装置,提高碳排放量估算的效率和准确性。
本申请提供了一种动力电池的年度碳排放量估算方法,包括:获取新能源汽车的年度销售数据,将所述年度销售数据输入到预训练的动力电池年度生产预测模型中,以使所述预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;将所述年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使所述预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;获取所述第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
在一种可能的实现方式中,获取所述第一重量动力电池预测生产量对应的第一碳排放量,包括:获取所述第一重量动力电池预测生产量中单个动力电池的所有组成材料,并获取每种组成材料对应的第一材料重量;根据所述第一材料重量和所述第一重量动力电池预测生产量,得到所述第一重量动力电池预测生产量中所述每种组成材料对应的第二材料重量;获取所述每种组成材料对应的标准单位碳排放量,基于所述第二材料重量和所述标准单位碳排放量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第一材料碳排放量,整合所有第一材料碳排放量,得到所述第一重量动力电池预测生产量对应的第一碳排放量。
在一种可能的实现方式中,动力电池年度生产预测模型的预训练过程,包括:获取不同年份对应的新能源汽车历史年度销售数据及动力电池历史年度生产量,得到基于时间序列分析的历史样本数据集;对所述历史样本数据集进行随机抽样处理,得到训练集;将所述训练集输入到神经网络进行前向传播和后向传播,得到动力电池年度生产预测模型。
在一种可能的实现方式中,动力电池重量分类模型的预训练过程,包括:获取不同年份对应的动力电池历史年度生产量,基于单个动力电池总重量的不同,对所述动力电池历史年度生产量进行分类,得到多种总重量对应的第一重量动力电池历史生产量;将所述动力电池历史年度生产量作为神经网络模型的输入,将所述多种总重量对应的第一重量动力电池历史生产量作为神经网络模型的输出,直至训练所述神经网络模型收敛,得到动力电池重量分类模型。
在一种可能的实现方式中,所述单个动力电池的所有组成材料包括正极材料、负极材料、电解液、隔膜、壳体和电池管理***BMS。
本申请还提供了一种动力电池的年度碳排放量估算装置,包括:年度动力电池预测生产量获取模块、动力电池重量分类模块和年度碳排放量获取模块;其中,所述年度动力电池预测生产量获取模块,设置为获取新能源汽车的年度销售数据,将所述年度销售数据输入到预训练的动力电池年度生产预测模型中,以使所述预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;所述动力电池重量分类模块,设置为将所述年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使所述预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;
所述年度碳排放量获取模块,设置为获取所述第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
在一种可能的实现方式中,所述年度碳排放量获取模块,设置为获取所述
第一重量动力电池预测生产量对应的第一碳排放量,包括:获取所述第一重量动力电池预测生产量中单个动力电池的所有组成材料,并获取每种组成材料对应的第一材料重量;根据所述第一材料重量和所述第一重量动力电池预测生产量,得到所述第一重量动力电池预测生产量中所述每种组成材料对应的第二材料重量;获取所述每种组成材料对应的标准单位碳排放量,基于所述第二材料重量和所述标准单位碳排放量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第一材料碳排放量,整合所有第一材料碳排放量,得到所述第一重量动力电池预测生产量对应的第一碳排放量。
在一种可能的实现方式中,所述年度动力电池预测生产量获取模块中动力电池年度生产预测模型的预训练过程,包括:获取不同年份对应的新能源汽车历史年度销售数据及动力电池历史年度生产量,得到基于时间序列分析的历史样本数据集;对所述历史样本数据集进行随机抽样处理,得到训练集;将所述训练集输入到神经网络进行前向传播和后向传播,得到动力电池年度生产预测模型。
在一种可能的实现方式中,所述动力电池重量分类模块中动力电池重量分类模型的预训练过程,包括:不同年份对应的动力电池历史年度生产量,基于单个动力电池总重量的不同,对所述动力电池历史年度生产量进行分类,得到多种总重量对应的第一重量动力电池历史生产量;所述动力电池历史年度生产量作为神经网络模型的输入,将所述多种总重量对应的第一重量动力电池历史生产量作为神经网络模型的输出,直至训练所述神经网络模型收敛,得到动力电池重量分类模型。
在一种可能的实现方式中,所述单个动力电池的所有组成材料包括正极材料、负极材料、电解液、隔膜、壳体和BMS。
本申请还提供了一种终端设备,包括处理器、存储器以及存储在所述存储器中且被配置为由所述处理器执行的计算机程序,所述处理器执行所述计算机程序时实现如上述任意一项所述的动力电池的年度碳排放量估算方法。
本申请还提供了一种计算机可读存储介质,所述计算机可读存储介质包括存储的计算机程序,其中,在所述计算机程序运行时控制所述计算机可读存储介质所在设备执行如上述任意一项所述的动力电池的年度碳排放量估算方法。
图1是本申请提供的一种动力电池的年度碳排放量估算方法的一种实施例的流程示意图;
图2是本申请提供的一种动力电池的年度碳排放量估算装置的一种实施例的结构示意图;
图3是本申请提供的终端设备的结构示意图。
下面将结合本申请中的附图,对本申请实施例中的技术方案进行描述,显然,所描述的实施例仅仅是本申请一部分实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
实施例1
参见图1,图1是本申请提供的一种动力电池的年度碳排放量估算方法的一种实施例的流程示意图,如图1所示,该方法包括步骤101-步骤103
步骤101:获取新能源汽车的年度销售数据,将所述年度销售数据输入到预训练的动力电池年度生产预测模型中,以使所述预训练的动力电池年度生产预测模型输出年度动力电池预测生产量。
一实施例中,对于动力电池年度生产预测模型,其预训练过程,通过获取不同年份对应的新能源汽车历史年度销售数据及动力电池历史年度生产量,得到基于时间序列分析的历史样本数据集,即对获取的新能源汽车历史年度销售数据及动力电池历史年度生产量按年份的先后顺序进行排列,得到数据序列,形成历史样本数据集。
可选的,依据数据获取的年份信息,对历史样本数据集中的每一个历史样本数据设置时间标签,其中,所述时间标签为年份数据。
一实施例中,对所述历史样本数据集进行随机抽样处理,得到训练集;将历史样本数据集按预设比例分为训练集和测试集,基于划分的训练集中历史样本数据的数量,从历史样本数据集中随机抽取相应的数量的历史样本数据,形成训练集,并将剩余部分作为测试集;其中,所述预设比例为8:2。
一实施例中,将所述训练集输入到神经网络模型进行前向传播和后向传播,得到动力电池年度生产预测模型。
将所述训练集输入到长短时记忆循环神经网络模型中进行训练,获取原始动力电池年度生产预测模型,其中,所述长短时记忆循环神经网络模型是一种时间递归神经网络模型,适合于处理和预测具有时间序列,且时间序列间隔和延迟相对较长的重要事件。
一实施例中,长短时记忆循环神经网络模型包括输入层、隐藏层和输出层三层网络结构。输入层是长短时记忆循环神经网络模型的第一层,用于接收外界信号,即负责接收训练集中的新能源汽车历史年度销售数据及动力电池历史年度生产量;输出层是长短时记忆循环神经网络模型的最后一层,用于向外界输出信号,即负责输出长短时记忆循环神经网络模型的计算预测结果;隐藏层是长短时记忆循环神经网络模型中除输入层和输出层之外的多个层,用于对训练集中的新能源汽车历史年度销售数据及动力电池历史年度生产量进行处理,获取长短时记忆循环神经网络模型的计算结果。可选的,采用长短时记忆循环神经网络模型进行模型训练增加了训练数据的时序性,从而提高了预测模型的准确率。
一实施例中,获取并基于前向传播算法对训练集进行训练,获取原始动力电池年度生产预测模型的第一状态参数,其中,第一状态参数是指基于训练数据进行模型训练时初始迭代过程所得到的参数;前向传播算法的计算公式为和其中,St表示t时刻隐藏层的输出;表示隐藏层上一时刻t-1到t时刻的权值;表示输入层到输出层的权值;表示t时刻的预测输出;表示隐藏层到所述输出层的权值。
一实施例中,获取并基于后向传播算法对所述第一状态参数进行误差计算,得到原始动力电池年度生产预测模型的第二状态参数,基于所述第二状态参数,得到原始动力电池年度生产预测模型。所述后向传播算法为式中,Ot表示t时刻与对应的真实值。
采用交叉熵损失函数计算出每一层的偏导数,基于所述偏导数来更新U、V和W这三个权值参数,以获取调节后的第二状态参数;其中,所述偏导数包括
一实施例中,在得到原始动力电池年度生产预测模型后,基于测试集,对所述原始动力电池年度生产预测模型进行测试,得到动力电池年度生产预测模型。
将测试集中的任一年份的新能源汽车历史年度销售数据依次输入到原始动力电池年度生产预测模型中,以使原始动力电池年度生产预测模型依次输出动力电池年度预测生产量,将所述动力电池年度预测生产量与该年份对应的所述动力电池历史年度生产量进行对比,得到预测误差值,判断所述预测误差值是否在预设误差范围内,若是,则将所述原始动力电池年度生产预测模型作为所述动力电池年度生产预测模型;若否,则将继续对所述原始动力电池年度生产预测模型进行训练,直至所述预测误差值在所述预测误差范围内。
步骤102:将所述年度动力电池预测生产量输入到预训练的动力电池重量分
类模型中,以使所述预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量。
一实施例中,动力电池的重量受动力电池的额定电压和容量的影响,不同的额定电压和容量,会导致动力电池的重量不同;因此基于新能源汽车的不同,其对应的动力电池重量也可能存在差异,基于此,还需要预构建动力电池重量分类模型,以使基于动力电池重量分类模型对预测得到的年度动力电池预测生产量按单个动力电池总重量的不同进行分类。
一实施例中,动力电池重量分类模型的训练过程,通过获取同年份对应的动力电池历史年度生产量,基于单个动力电池总重量的不同,对所述动力电池历史年度生产量进行分类,得到多种总重量对应的第一重量动力电池历史生产量;将所述动力电池历史年度生产量作为神经网络模型的输入,将所述多种总重量对应的第一重量动力电池历史生产量作为神经网络模型的输出,直至训练所述神经网络模型收敛,得到动力电池重量分类模型。
基于获取的不同年份对应的动力电池历史年度生产量,生成历史生产量数据集,将所述历史生产量数据集按预设比例划分为第一训练集和第一测试集;其中,预测比例为8:2。
一实施例中,基于第一训练集对神经网络模型进行训练,并通过记录模型训练过程中的损失值和准确度值,确定原始动力电池重量分类模型。
通过TensorBoard可视化技术,使模型训练直观可控,同时将每步训练得出的损失值和准确度值记录下来,然后每隔5步用测试集进行模型准确度计算,得出验证的损失值和准确度值,并将验证的损失值和准确度值记录在TensorBoard中,以使根据将损失值和准确度值绘制成变化曲线图,用于衡量当前训练进度下的模型识别效果。
在进行模型训练时,还设置了损失变化阈值和最大迭代次数,通过损失变化阈值和最大迭代次数判断是否停止对模型进行训练;当模型训练未达到最大迭代次数,且损失值的变化在一定的周期内一直没有达到损失变化阈值时,判断验证的准确度值是否开始下降时,若否,则继续对模型进行训练,若是,则说明此时模型的拟合度达到最优的峰值,接下来若再训练就会使验证数据集的准确度持续下降,导致模型过拟合,所以此时应停止训练。
一实施例中,基于第一测试集对所述原始动力电池重量分类模型进行测试,得到动力电池重量分类模型。
将第一测试集中的任一年份的动力电池历史年度生产量依次输入到原始动力电池重量分类模型中,以使原始动力电池重量分类模型依次输出多种总重量
对应的第一重量动力电池历史预测生产量,将第一重量动力电池历史预测生产量与该年份对应的所述第一重量动力电池历史生产量进行对比,得到第一预测误差值,判断所述第一预测误差值是否在第一预设误差范围内,若是,则将所述原始动力电池重量分类模型作为动力电池重量分类模型;若否,则将继续对所述原始动力电池重量分类模型进行训练,直至所述第一预测误差值在所述第一预测误差范围内。
步骤103:获取所述第一重量动力电池预设生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
一实施例中,获取所述第一重量动力电池预测生产量中单个动力电池的所有组成材料,并获取每种组成材料对应的第一材料重量。
一实施例中,所述单个动力电池的所有组成材料包括正极材料、负极材料、电解液、隔膜、壳体和电池管理***(Battery Management System,BMS)。
可选的,所述正极材料包括磷酸铁锂、聚偏氟乙烯、N-甲基吡咯烷酮和铝箔;所述负极材料包括石墨、聚偏氟乙烯、N-甲基吡咯烷酮、铜箔;所述电解液包括六氟磷酸锂、碳酸乙烯酯和碳酸二甲酯;所述隔膜包括聚丙烯和聚乙烯;所述壳体包括铝片;所述BMS包括电路板、钢片和铜片。
一实施例中,获取每种组成材料在单个动力电池中对应的第一材料重量,通过获取每种组成材料对应的标准单位所需材料质量,基于单个动力电池总重量和所述标准单位所需材料质量,得到每种组成材料在单个动力电池中对应的第一材料重量。计算过程过下所示:mi=Pi×m总;其中,mi为第i种所需材料在单个动力电池中的第一材料重量,单位为kg,m总为年度动力电池预测生产量的总重量,单位为kg,Pi为第i种组成材料对应的标准单位所需材料质量,单位为kg。
一实施例中,根据所述第一材料重量和所述第一重量动力电池预测生产量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第二材料重量。
由于第一材料重量为单个动力电池中的单个所需组成材料的重量,因此将所述第一材料重量乘以该单个动力电池总重量对应的第一重量动力电池预测生产量,就能得到该种所需组成材料在第一重量动力电池预测生产量的总重量,即第二材料重量;计算过程如下所示:Mi=Si×mi;其中,Mi为第i种所需材料在第一重量动力电池预测生产量中的第二材料重量,单位为kg,Si为第一重量动力电池预测生产量。
一实施例中,获取每种组成材料对应的标准单位碳排放量,基于所述第二材料重量和所述标准单位碳排放量,得到所述第一重量动力电池预测生产量中
每种组成材料对应的第一材料碳排放量,整合所有第一材料碳排放量,得到所述第一重量动力电池预测生产量对应的第一碳排放量。
对于第一材料碳排放量的计算过程,如下所示:Ei=Mi×ei;其中,Ei为第i种所需组成材料对应的第一材料碳排放量,单位为kg,ei为第i种所需组成材料对应的标准单位碳排放量。
对计算出来的所有所需组成材料对应的第一材料碳排放量进行叠加处理,形成所述第一重量动力电池预测生产量对应的第一碳排放量,计算过程如下所示:其中,wt为第t组第一重量动力电池预测生产量对应的第一碳排放量。
一实施例中,整合所有第一碳排放量,得到年度碳排放量;对所有第一碳排放量进行叠加处理,形成所述年度动力电池预测生产量的年度碳排放量,实现对年度碳排放量的预测;其中年度碳排放量的计算过程如下所示:其中,W为年度碳排放量,单位为kg。
一实施例中,还可基于预测未来目标年度对应的年度碳排放量,将所述未来目标年度对应的年度碳排放量与当前年度对应的年度碳排放量进行对比,得到年度碳排放量增长趋势,其中,所述年度碳排放量增长趋势可以为上升或下降,以使相关企业能及时了解到碳排放相关信息,对企业后续开展碳排放处理提供重要的绿色支撑。
实施例2
参见图2,图2是本申请提供的一种动力电池的年度碳排放量估算装置的一种实施例的结构示意图,如图2所示,该装置包括年度动力电池预测生产量获取模块201、动力电池重量分类模块202和年度碳排放量获取模块203。
所述年度动力电池预测生产量获取模块201,设置为获取新能源汽车的年度销售数据,将所述年度销售数据输入到预训练的动力电池年度生产预测模型中,以使所述预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;所述动力电池重量分类模块202,设置为将所述年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使所述预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;所述年度碳排放量获取模块203,设置为获取所述第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
一实施例中,所述年度碳排放量获取模块203,设置为获取所述第一重量动力电池预测生产量对应的第一碳排放量,包括:获取所述第一重量动力电池预测生产量中单个动力电池的所有组成材料,并获取每种组成材料对应的第一材
料重量;根据所述第一材料重量和所述第一重量动力电池预测生产量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第二材料重量;获取每种组成材料对应的标准单位碳排放量,基于所述第二材料重量和所述标准单位碳排放量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第一材料碳排放量,整合所有第一材料碳排放量,得到所述第一重量动力电池预测生产量对应的第一碳排放量。
一实施例中,所述年度动力电池预测生产量获取模块201中动力电池年度生产预测模型的预训练过程,包括:获取不同年份对应的新能源汽车历史年度销售数据及动力电池历史年度生产量,得到基于时间序列分析的历史样本数据集;对所述历史样本数据集进行随机抽样处理,得到训练集;将所述训练集输入到神经网络进行前向传播和后向传播,得到动力电池年度生产预测模型。
一实施例中,所述动力电池重量分类模块202中动力电池重量分类模型的预训练过程,包括:获取不同年份对应的动力电池历史年度生产量,基于单个动力电池总重量的不同,对所述动力电池历史年度生产量进行分类,得到多种总重量对应的第一重量动力电池历史生产量;将所述动力电池历史年度生产量作为神经网络模型的输入,将所述多种总重量对应的第一重量动力电池历史生产量作为神经网络模型的输出,直至训练所述神经网络模型收敛,得到动力电池重量分类模型。
一实施例中,所述单个动力电池的所有组成材料包括正极材料、负极材料、电解液、隔膜、壳体和BMS。
所属领域的技术人员可以清楚的了解到,为描述的方便和简洁,上述描述的装置的工作过程,可以参考前述方法实施例中的对应过程,在此不在赘述。
需要说明的是,上述动力电池的年度碳排放量估算装置的实施例仅仅是示意性的,其中所述作为分离部件说明的模块可以是或者也可以不是物理上分开的,作为模块显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案。
在上述的动力电池的年度碳排放量估算方法的实施例的基础上,本申请另一实施例提供了一种动力电池的年度碳排放量估算终端设备,该动力电池的年度碳排放量估算终端设备,包括处理器、存储器以及存储在所述存储器中且被配置为由所述处理器执行的计算机程序,所述处理器执行所述计算机程序时,实现本申请任意一实施例的动力电池的年度碳排放量估算方法。
示例性的,在这一实施例中所述计算机程序可以被分割成一个或多个模块,
所述一个或者多个模块被存储在所述存储器中,并由所述处理器执行,以完成本申请。所述一个或多个模块可以是能够完成特定功能的一系列计算机程序指令段,该指令段用于描述所述计算机程序在所述动力电池的年度碳排放量估算终端设备中的执行过程。
所述动力电池的年度碳排放量估算终端设备可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述动力电池的年度碳排放量估算终端设备可包括,但不仅限于,处理器、存储器。
所称处理器可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等,所述处理器是所述动力电池的年度碳排放量估算终端设备的控制中心,利用多种接口和线路连接整个动力电池的年度碳排放量估算终端设备的多个部分。
所述存储器可设置为存储所述计算机程序和/或模块,所述处理器通过运行或执行存储在所述存储器内的计算机程序和/或模块,以及调用存储在存储器内的数据,实现所述动力电池的年度碳排放量估算终端设备的多种功能。所述存储器可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作***、至少一个功能所需的应用程序等;存储数据区可存储根据手机的使用所创建的数据等。此外,存储器可以包括高速随机存取存储器,还可以包括非易失性存储器,例如硬盘、内存、插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
在上述动力电池的年度碳排放量估算方法的实施例的基础上,本申请另一实施例提供了一种存储介质,所述存储介质包括存储的计算机程序,其中,在所述计算机程序运行时,控制所述存储介质所在的设备执行本申请任意一实施例的动力电池的年度碳排放量估算方法。
在这一实施例中,上述存储介质为计算机可读存储介质,所述计算机程序包括计算机程序代码,所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或一些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、电载波信号、电信信号以及软件分发介质
等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在一些司法管辖区,根据立法和专利实践,计算机可读介质不包括电载波信号和电信信号。
综上,本申请提供的一种动力电池的年度碳排放量估算方法及装置,通过获取的新能源汽车的年度销售数据输入到预训练的动力电池年度生产预测模型中,以使预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;将年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;获取第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。本申请帮助企业基于新能源汽车的年度销售量来预测动力电池的生产量,并基于动力电池重量的不同,对预测的动力电池生产量进行分类,能提高碳排放量估算的效率和准确性。
Claims (10)
- 一种动力电池的年度碳排放量估算方法,包括:获取新能源汽车的年度销售数据,将所述年度销售数据输入到预训练的动力电池年度生产预测模型中,以使所述预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;将所述年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使所述预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;获取所述第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
- 根据权利要求1所述的方法,其中,获取所述第一重量动力电池预测生产量对应的第一碳排放量,包括:获取所述第一重量动力电池预测生产量中单个动力电池的所有组成材料,并获取每种组成材料对应的第一材料重量;根据所述第一材料重量和所述第一重量动力电池预测生产量,得到所述第一重量动力电池预测生产量中所述每种组成材料对应的第二材料重量;获取所述每种组成材料对应的标准单位碳排放量,基于所述第二材料重量和所述标准单位碳排放量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第一材料碳排放量,整合所有第一材料碳排放量,得到所述第一重量动力电池预测生产量对应的第一碳排放量。
- 根据权利要求1所述的方法,其中,动力电池年度生产预测模型的预训练过程,包括:获取不同年份对应的新能源汽车历史年度销售数据及动力电池历史年度生产量,得到基于时间序列分析的历史样本数据集;对所述历史样本数据集进行随机抽样处理,得到训练集;将所述训练集输入到神经网络进行前向传播和后向传播,得到动力电池年度生产预测模型。
- 根据权利要求1所述的方法,其中,动力电池重量分类模型的预训练过程,包括:获取不同年份对应的动力电池历史年度生产量,基于单个动力电池总重量的不同,对所述动力电池历史年度生产量进行分类,得到多种总重量对应的第一重量动力电池历史生产量;将所述动力电池历史年度生产量作为神经网络模型的输入,将所述多种总重量对应的第一重量动力电池历史生产量作为神经网络模型的输出,直至训练所述神经网络模型收敛,得到动力电池重量分类模型。
- 根据权利要求2所述的方法,其中,所述单个动力电池的所有组成材料包括正极材料、负极材料、电解液、隔膜、壳体和电池管理***BMS。
- 一种动力电池的年度碳排放量估算装置,包括:年度动力电池预测生产量获取模块、动力电池重量分类模块和年度碳排放量获取模块;其中,所述年度动力电池预测生产量获取模块,设置为获取新能源汽车的年度销售数据,将所述年度销售数据输入到预训练的动力电池年度生产预测模型中,以使所述预训练的动力电池年度生产预测模型输出年度动力电池预测生产量;所述动力电池重量分类模块,设置为将所述年度动力电池预测生产量输入到预训练的动力电池重量分类模型中,以使所述预训练的动力电池重量分类模型基于单个动力电池总重量的不同,输出多种总重量对应的第一重量动力电池预测生产量;所述年度碳排放量获取模块,设置为获取所述第一重量动力电池预测生产量对应的第一碳排放量,整合所有第一碳排放量,得到年度碳排放量。
- 根据权利要求6所述的装置,其中,所述年度碳排放量获取模块,设置为获取所述第一重量动力电池预测生产量对应的第一碳排放量,包括:获取所述第一重量动力电池预测生产量中单个动力电池的所有组成材料,并获取每种组成材料对应的第一材料重量;根据所述第一材料重量和所述第一重量动力电池预测生产量,得到所述第一重量动力电池预测生产量中所述每种组成材料对应的第二材料重量;获取所述每种组成材料对应的标准单位碳排放量,基于所述第二材料重量和所述标准单位碳排放量,得到所述第一重量动力电池预测生产量中每种组成材料对应的第一材料碳排放量,整合所有第一材料碳排放量,得到所述第一重量动力电池预测生产量对应的第一碳排放量。
- 根据权利要求6所述的装置,其中,所述年度动力电池预测生产量获取模块中动力电池年度生产预测模型的预训练过程,包括:获取不同年份对应的新能源汽车历史年度销售数据及动力电池历史年度生产量,得到基于时间序列分析的历史样本数据集;对所述历史样本数据集进行随机抽样处理,得到训练集;将所述训练集输入到神经网络进行前向传播和后向传播,得到动力电池年度生产预测模型。
- 一种终端设备,包括处理器、存储器以及存储在所述存储器中且被配置为由所述处理器执行的计算机程序,所述处理器执行所述计算机程序时实现如权利要求1至5任意一项所述的动力电池的年度碳排放量估算方法。
- 一种计算机可读存储介质,包括存储的计算机程序,其中,在所述计算机程序运行时控制所述计算机可读存储介质所在设备执行如权利要求1至5中任意一项所述的动力电池的年度碳排放量估算方法。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211177780.8A CN115600728A (zh) | 2022-09-23 | 2022-09-23 | 一种动力电池的年度碳排放量估算方法及装置 |
CN202211177780.8 | 2022-09-23 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024060539A1 true WO2024060539A1 (zh) | 2024-03-28 |
Family
ID=84845016
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2023/081880 WO2024060539A1 (zh) | 2022-09-23 | 2023-03-16 | 动力电池的年度碳排放量估算方法及装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN115600728A (zh) |
WO (1) | WO2024060539A1 (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117236750A (zh) * | 2023-09-08 | 2023-12-15 | 重庆环问问科技有限公司 | 一种基于差异数据分析的全周期碳排放计算方法 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115600728A (zh) * | 2022-09-23 | 2023-01-13 | 广东邦普循环科技有限公司(Cn) | 一种动力电池的年度碳排放量估算方法及装置 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108320064A (zh) * | 2018-04-28 | 2018-07-24 | 国电南瑞南京控制***有限公司 | 一种电动汽车与风电协同充电双层优化调度方法 |
CN113239314A (zh) * | 2021-04-09 | 2021-08-10 | 国网河北省电力有限公司沧州供电分公司 | 用于碳排放量预测的方法、装置、终端及计算机可读存储介质 |
CN114004634A (zh) * | 2021-10-18 | 2022-02-01 | 广东邦普循环科技有限公司 | 一种全球动力电池报废量的预测方法及装置 |
CN114090939A (zh) * | 2021-11-05 | 2022-02-25 | 国网湖北省电力有限公司经济技术研究院 | 一种储能电池全生命周期碳排放系数确定方法 |
CN115600728A (zh) * | 2022-09-23 | 2023-01-13 | 广东邦普循环科技有限公司(Cn) | 一种动力电池的年度碳排放量估算方法及装置 |
-
2022
- 2022-09-23 CN CN202211177780.8A patent/CN115600728A/zh active Pending
-
2023
- 2023-03-16 WO PCT/CN2023/081880 patent/WO2024060539A1/zh unknown
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108320064A (zh) * | 2018-04-28 | 2018-07-24 | 国电南瑞南京控制***有限公司 | 一种电动汽车与风电协同充电双层优化调度方法 |
CN113239314A (zh) * | 2021-04-09 | 2021-08-10 | 国网河北省电力有限公司沧州供电分公司 | 用于碳排放量预测的方法、装置、终端及计算机可读存储介质 |
CN114004634A (zh) * | 2021-10-18 | 2022-02-01 | 广东邦普循环科技有限公司 | 一种全球动力电池报废量的预测方法及装置 |
CN114090939A (zh) * | 2021-11-05 | 2022-02-25 | 国网湖北省电力有限公司经济技术研究院 | 一种储能电池全生命周期碳排放系数确定方法 |
CN115600728A (zh) * | 2022-09-23 | 2023-01-13 | 广东邦普循环科技有限公司(Cn) | 一种动力电池的年度碳排放量估算方法及装置 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117236750A (zh) * | 2023-09-08 | 2023-12-15 | 重庆环问问科技有限公司 | 一种基于差异数据分析的全周期碳排放计算方法 |
CN117236750B (zh) * | 2023-09-08 | 2024-07-05 | 重庆环问问科技有限公司 | 一种基于差异数据分析的全周期碳排放计算方法 |
Also Published As
Publication number | Publication date |
---|---|
CN115600728A (zh) | 2023-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2024060539A1 (zh) | 动力电池的年度碳排放量估算方法及装置 | |
Ma et al. | A novel method for state of health estimation of lithium-ion batteries based on improved LSTM and health indicators extraction | |
Xu et al. | Improving the state of charge estimation of reused lithium-ion batteries by abating hysteresis using machine learning technique | |
KR102439041B1 (ko) | 신경망 기반의 배터리 셀 불량 및 화재 사전 진단 방법 및 장치 | |
CN110412470B (zh) | 电动汽车动力电池soc估计方法 | |
CN110658460B (zh) | 一种电池包的电池寿命预测方法及装置 | |
CN110806541B (zh) | 一种基于ad-bas的锂电池模型参数辨识方法 | |
Jiang et al. | State of health estimation of lithium-ion battery with automatic feature extraction and self-attention learning mechanism | |
CN111231758B (zh) | 一种电池容量的估算方法、装置、电子设备及介质 | |
CN116609676B (zh) | 一种基于大数据处理的混合储能电池状态监控方法及*** | |
Tian et al. | Data‐driven battery degradation prediction: Forecasting voltage‐capacity curves using one‐cycle data | |
CN115718263B (zh) | 基于注意力的锂离子电池日历老化预测模型和方法 | |
CN112686380A (zh) | 基于神经网络的梯次动力电芯一致性评估方法和*** | |
CN116400241A (zh) | 基于全生命周期运行数据的电池健康评估方法及装置 | |
Tan et al. | Intelligent online health estimation for lithium-ion batteries based on a parallel attention network combining multivariate time series | |
CN115407211A (zh) | 一种电动汽车锂电池健康状态在线预测方法及*** | |
Liu et al. | Dual particle swarm optimization based data-driven state of health estimation method for lithium-ion battery | |
Wang et al. | A novel capacity estimation method for li-ion battery cell by applying ensemble learning to extremely sparse significant points | |
CN112084459A (zh) | 电池充放电循环寿命的预测方法、装置、电子终端、及存储介质 | |
CN117289167A (zh) | 基于多重神经网络的电池剩余寿命预测方法、装置及介质 | |
Zhang et al. | Remaining useful life prediction of lithium-ion batteries based on TCN-DCN fusion model combined with IRRS filtering | |
CN115407217B (zh) | 一种电动汽车锂电池荷电状态在线估计方法及*** | |
CN116008815A (zh) | 一种电芯内短路检测方法、装置、设备、存储介质及车辆 | |
WO2024093005A1 (zh) | 一种电池容量预测方法、装置及电子设备 | |
US20220146996A1 (en) | Reward generation method to reduce peak load of electric power and action control apparatus performing the same method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23866840 Country of ref document: EP Kind code of ref document: A1 |