METHOD,APPARATUS,DEVICE,AND MEDIUM FOR PREDICTING CARBON EMMSIONS FROM INDUSTRIAL PROCESS
FIELD
-
The present disclosure relates to the technical field of carbon emissions technology, in particular to a method, apparatus, device, and medium for predicting carbon emissions from an industrial process.
BACKGROUND
-
The carbon footprint consulting service of industrial products has become an emerging hot spot in the industry. Carbon emissions refer to the greenhouse gas emissions generated during the production, transportation, use and recycling of products.
-
Generally, in view of different workflows of industrial processes (such as production lines) , the carbon emissions calculation process is usually not reusable. Experts need to spend a lot of time to calculate carbon emissions of each industrial process, and the calculation process also requires high software and hardware costs.
SUMMARY
-
Embodiments of the present disclosure propose a method, apparatus, electronic device and medium for predicting carbon emissions from an industrial process.
-
In a first aspect, a method for predicting carbon emissions from an industrial process is provided. The method comprising:
-
determining the number of types of devices comprised in an industrial process and the number of devices of each type;
-
obtaining prediction values of carbon emissions for respective types of devices in the industrial process, wherein the prediction values of carbon emissions for respective types of devices are output from respective prediction models of carbon emissions associated with respective types of devices, and each prediction model of carbon emissions is determined through AI (artificial intelligence) ; and
-
determining prediction value of carbon emissions of the industrial process based on the number of types, the number of devices of each type and the prediction values for respective types of devices in the industrial process.
-
Therefore, embodiments of the present disclosure introduce AI capability into the process of carbon emissions prediction, reducing workload of carbon emissions prediction. In addition, devices with the same type share a carbon emission prediction model, there is no need to build special model for each device, reducing
implementation difficulty of carbon emissions prediction.
-
Preferably, wherein determining the number of types of devices comprised in an industrial process and the number of devices of each type comprises:
-
selecting devices in the industrial process from a device-list area in a visual interface in response to a selection command;
-
combining the selected devices into a target area in the visual interface to form the industrial process in response to a combination command;
-
counting the number of types of devices and the number of devices of each type in the target area.
-
Therefore, the number of types of devices and the number of devices of each type in the industrial process can be easily determined through visualization.
-
Preferably, comprising:
-
obtaining operation data of a device of target type during a historical execution period of the industrial process;
-
obtaining an actual value of carbon emissions of the device of target type during the historical execution period of the industrial process;
-
determining a training data set comprising the operation data and the actual value of carbon emissions as a label;
-
inputting the training data set into a neural network model, wherein the neural network model is configured to output a prediction value of carbon emissions based on the operation data;
-
determining a loss function value of the neural network model based on the difference between the prediction value of carbon emissions and the actual value of carbon emissions;
-
configuring model parameters of the neural network model to make the loss function value lower than a preset threshold;
-
determining configured neural network model as prediction model of the target type.
-
Therefore, based on historical operation data of each type of device and actual value of carbon emissions of the type of device, it is convenient to train prediction model of carbon emissions for each type of devices.
-
Preferably, wherein obtaining operation data of device of target type during a historical execution period of the industrial process comprises:
-
deploying a detachable electricity meter for the device of target type before the historical execution period of the industrial process;
-
acquiring a reading of the electricity meter during the historical execution period of the industrial process;
-
converting the reading into the actual value of carbon emissions;
-
removing the detachable electricity meter from the device.
-
Therefore, the actual value of carbon emissions of the type of device during the historical execution period can be easily determined through detachable electricity meter.
-
Preferably, wherein the operation data comprises a general data type and a special data type, the general data type is applicable to all types of devices in the industrial process, and the special data type is applicable to the target type of device.
-
Therefore, operation data can be implemented in many ways, which not only reflects individual attributes of carbon emissions of specific device, but also reflects common attributes of carbon emissions between different devices.
-
Preferably, wherein the general data type comprises at least one of the following:
-
device utilization; operation cycle; idle time.
-
Therefore, common data types have many implementations.
-
Preferably, wherein determining prediction value of carbon emissions of the industrial process based on the number of types, the number of devices of each type and the prediction values for respective types of devices in the industrial process comprises:
-
determining respective prediction values of carbon emissions of all devices of respective same types based on respective numbers of devices of respective same types and respective prediction values output from respective prediction models of carbon emissions associated with respective same types;
-
calculating sum of respective prediction values of carbon emissions of all devices of respective same types;
-
determining the sum as the prediction value of carbon emissions of the industrial process.
-
Therefore, by classifying the devices in the industrial process according to respective device types, and then summing up based on the classification results, prediction value of carbon emissions of the entire industrial process can be easily determined.
-
In a second aspect, an apparatus for predicting carbon emissions from an industrial process is provided. The apparatus comprising:
-
a first determining module, configured to determine the number of types of devices comprised in an industrial process and the number of devices of each type;
-
an obtaining module, configured to obtain prediction values of carbon emissions for respective types of devices in the industrial process, wherein the prediction values of carbon emissions for respective types of devices are output from respective prediction models of carbon emissions associated with respective types of devices, and
each prediction model of carbon emissions is determined through artificial intelligence; and
-
a second determining module, configured to determine prediction value of carbon emissions of the industrial process based on the number of types, the number of devices of each type and the prediction values for respective types of devices in the industrial process.
-
Therefore, embodiments of the present disclosure introduce AI capability into the process of carbon emissions prediction, reducing workload of carbon emissions prediction. In addition, devices with the same type share a carbon emission prediction model, there is no need to build special model for each device, reducing implementation difficulty of carbon emissions prediction.
-
In a third aspect, a method for determining a prediction model of carbon emissions, wherein the prediction model corresponds to a type of devices comprised in an industrial process; the prediction model is adapted to output prediction value of carbon emissions of the corresponding type of devices in the industrial process; the prediction model is determined by artificial intelligence; and prediction value of carbon emissions of the industrial process is determined based on the number of types of devices comprised in the industrial process, the number of devices of each type and prediction values for respective types of devices in the industrial process; the method comprises:
-
obtaining operation data of a device comprised in the industrial process during a historical execution period of the industrial process;
-
obtaining an actual value of carbon emissions of the device during the historical execution period of the industrial process;
-
determining a training data set comprising the operation data and the actual value of carbon emissions as a label;
-
inputting the training data set into a neural network model, wherein the neural network model is configured to output a prediction value of carbon emissions based on the operation data;
-
determining a loss function value of the neural network model based on the difference between the prediction value of carbon emissions and the actual value of carbon emissions;
-
configuring model parameters of the neural network model to make the loss function value lower than a preset threshold; and
-
determining the configured neural network model as prediction model of the type of the device.
-
Preferably, wherein obtaining operation data of a device comprised in the industrial process during a historical execution period of the industrial process comprises:
-
deploying a detachable electricity meter for the device before the historical execution period of the industrial
process;
-
acquiring a reading of the electricity meter during the historical execution period of the industrial process;
-
converting the reading into the actual value of carbon emissions;
-
removing the detachable electricity meter from the device.
-
Preferably, wherein the operation data comprises a general data type and a special data type, the general data type is applicable to all types of devices in the industrial process, and the special data type is applicable to the type of the device.
-
In a fourth aspect, an electronic device is provided. The electronic device comprising a processor and a memory, wherein an application program executable by the processor is stored in the memory for causing the processor to execute a method for predicting carbon emissions from an industrial process as described in any of the above or a method for determining a prediction model of carbon emissions as described in any of the above.
-
In a fifth aspect, a computer-readable medium comprising computer-readable instructions stored thereon is provided, wherein the computer-readable instructions, when executed by a processor, implement a method for building knowledge graph or a method for predicting carbon emissions from an industrial process as described in any of the above or a method for determining a prediction model of carbon emissions as described in any of the above.
-
In a sixth aspect, a computer program product comprising a computer program, when the computer program is executed by a processor for executing a method for predicting carbon emissions from an industrial process as described in any of the above or a method for determining a prediction model of carbon emissions as described in any of the above.
BRIEF DESCRIPTION OF THE DRAWINGS
-
In order to make technical solutions of examples of the present disclosure clearer, accompanying drawings to be used in description of the examples will be simply introduced hereinafter. Obviously, the accompanying drawings to be described hereinafter are only some examples of the present disclosure. Those skilled in the art may obtain other drawings according to these accompanying drawings without creative labor.
-
FIG. 1 is a flowchart of a method for determining a prediction model of carbon emissions according to embodiments of the present disclosure.
-
FIG. 2 is a schematic diagram of training process of prediction model according to embodiments of the present disclosure.
-
FIG. 3 is a schematic diagram of predicting carbon emissions of a device based on prediction model
according to embodiments of the disclosure.
-
FIG. 4 is a flowchart of a method for predicting carbon emissions from an industrial process according to embodiments of the present disclosure.
-
FIG. 5 is a schematic diagram of the carbon emissions prediction process of industrial process according to embodiments of the present disclosure.
-
FIG. 6 is a structural diagram of apparatus for predicting carbon emissions from an industrial process according to embodiments of the present disclosure.
-
FIG. 7 is a structural diagram of an electronic device according to embodiments of the present disclosure.
-
List of reference numbers:
DETAILED DESCRIPTION
-
In order to make the purpose, technical scheme, and advantages of the disclosure clearer, the following examples are given to further explain the disclosure in detail.
-
In order to be concise and intuitive in description, the scheme of the disclosure is described below by describing several representative embodiments. Many details in the embodiments are only used to help understand the scheme of the disclosure. However, it is obvious that the technical scheme of the disclosure can be realized without being limited to these details. In order to avoid unnecessarily blurring the scheme of the disclosure, some embodiments are not described in detail, but only the framework is given. Hereinafter, "including" refers to "including but not limited to" , "according to... " refers to "at least according to..., but not limited to... " . Due to the language habits of Chinese, when the number of an element is not specifically indicated below, it means that the element can be one or more, or can be understood as at least one.
-
According to embodiments of the present disclosure, respective carbon emissions prediction models are established according to respective types of devices in an industrial process, to quickly estimate carbon emissions from the whole industrial process. The carbon emissions of different types of devices are calculated in the unit of device in the production process, rather than in the unit of production process (such as production line or section) . By combining different types of devices and the number of devices of each type, different industrial processes can be quickly simulated, thus facilitating the prediction of carbon emissions in different industrial processes. In addition, the carbon emissions prediction model can be continuously calibrated in industrial environment by adding a detachable electricity meter to key energy consumption devices in the industrial process, to ensure the accuracy of the prediction value as much as possible while rapidly simulating.
-
FIG. 1 is a flowchart of a method for determining a prediction model of carbon emissions according to embodiments of the present disclosure. The prediction model corresponds to a type of devices comprised in an industrial process; the prediction model is adapted to output prediction value of carbon emissions of the corresponding type of devices in the industrial process; the prediction model is determined by AI. Prediction value of carbon emissions of the industrial process is determined based on the number of types of devices comprised in the industrial process, the number of devices of each type and prediction values for respective types of devices in the industrial process. As shown in Figure 1, the method comprises:
-
Step 101: obtaining operation data of a device comprised in the industrial process during a historical
execution period of the industrial process.
-
Step 102: obtaining an actual value of carbon emissions of the device during the historical execution period of the industrial process.
-
Step 103: determining a training data set comprising the operation data and the actual value of carbon emissions as a label.
-
Step 104: inputting the training data set into a neural network model, wherein the neural network model is configured to output a prediction value of carbon emissions based on the operation data.
-
Step 105: determining a loss function value of the neural network model based on the difference between the prediction value of carbon emissions and the actual value of carbon emissions.
-
Step 106: configuring model parameters of the neural network model to make the loss function value lower than a preset threshold.
-
Step 107: determining the configured neural network model as prediction model of the type of the device.
-
The process shown in Figure 1 is a general process of training respective carbon emission prediction models for respective types of devices. Based on the process shown in Figure 1, respective carbon emission prediction models for respective device types in the industrial process can be generated.
-
In one embodiment, wherein obtaining operation data of a device comprised in the industrial process during a historical execution period of the industrial process comprises: deploying a detachable electricity meter for the device before the historical execution period of the industrial process; acquiring a reading of the electricity meter during the historical execution period of the industrial process; converting the reading into the actual value of carbon emissions; removing the detachable electricity meter from the device.
-
In one embodiment, wherein the operation data comprises a general data type and a special data type, the general data type is applicable to all types of devices in the industrial process, and the special data type is applicable to the type of the device.
-
FIG. 2 is a schematic diagram of training process of prediction model according to embodiments of the present disclosure.
-
As shown in Figure 2, training data set 23 corresponding to device type 22 is determined. The training data set 23 contains operation data of device type 22 during historical execution period of the industrial process. When there are multiple devices conforming to device type 22 in the industrial process, operation data of one device can be arbitrarily selected as the operation data of device type 22. A detachable electricity meter is arranged for the selected device. Acquire the electricity meter reading 21 during the historical execution period of the industrial process. The meter reading 21 is added to the training data set 23 as a label of the operation data of the device type
22. Input training data set 23 into neural network model 24. The neural network model 24 outputs prediction of carbon emissions 26 based on the operation data of the device type 22. The loss function value of neural network model 24 is determined based on the difference between the prediction of carbon emissions 26 and the electricity meter reading 21. Configure model parameters of neural network model 24 to make the loss function value lower than a preset threshold, thus completing the training process of neural network model 23.
-
So far, the neural network model 23 that completes the training process, namely the carbon emission prediction model of device type 22 has been obtained. All devices of the same type can share the carbon emission prediction model of this type of device without special modeling for each device, which reduces the implementation difficulty of carbon emissions prediction.
-
FIG. 3 is a schematic diagram of predicting carbon emissions of a device based on prediction model according to embodiments of the disclosure. Input real-time operation data 27 of a certain device of device type 22 into the neural network model 24 that completes the training process. The neural network model 24 that completes the training process outputs: real-time prediction of carbon emissions 28 of the device in the industrial process.
-
FIG. 4 is a flowchart of a method for predicting carbon emissions from an industrial process according to embodiments of the present disclosure. As shown in Figure 4, the method comprises:
-
Step 201: determining the number of types of devices comprised in an industrial process and the number of devices of each type.
-
Here, industrial process refers to the production process in the production unit through physical and chemical changes. The production unit refers to the unit that can engage in production and business activities, including industry, mining, commerce, and trade, etc. Specifically, production units can be implemented as manufacturing groups, manufacturing plants, production workshops, production lines or assembly lines, etc. Industrial processes can include many types of devices, and the number of each type of device can be one or more. For example, a set of bottled mineral water production line can be composed of a variety of devices, including: bottle blowing machine, bottle management machine, water treatment system, filling machine, light inspection equipment, drying machine, laser inkjet printer, packing machine and stacking machine, etc.
-
In one embodiment, the number of types of devices in the industrial process and the number of devices of each type can be directly obtained from user input. Optionally, the number of types of devices and the number of devices of each type can be determined based on user's trigger instructions (such as: select instructions and drag instructions, etc. ) in a visual interface.
-
In one embodiment, wherein determining the number of types of devices comprised in an industrial process
and the number of devices of each type comprises: selecting devices in the industrial process from a device-list area in a visual interface in response to a selection command; combining the selected devices into a target area in the visual interface to form the industrial process in response to a combination command (for example, drag operation sent by the user) ; counting the number of types of devices and the number of devices of each type in the target area. Therefore, the number of types of devices and the number of devices of each type in the industrial process can be easily determined through visualization.
-
Step 202: obtaining prediction values of carbon emissions for respective types of devices in the industrial process, wherein the prediction values of carbon emissions for respective types of devices are output from respective prediction models of carbon emissions associated with respective types of devices, and each prediction model of carbon emissions is determined through artificial intelligence.
-
In one embodiment, it also includes a specific process of determining each type of carbon emission prediction model in the industrial process in AI mode. Any device type that needs to be generated a carbon emission prediction model can be called target type. That is, the target type can be any device type in the industrial process.
-
In one embodiment, process of determining carbon emission prediction model of each type of device includes: obtaining operation data of a device of target type during a historical execution period of the industrial process; obtaining an actual value of carbon emissions of the device of target type during the historical execution period of the industrial process; determining a training data set comprising the operation data and the actual value of carbon emissions as a label; inputting the training data set into a neural network model, wherein the neural network model is configured to output a prediction value of carbon emissions based on the operation data; determining a loss function value of the neural network model based on the difference between the prediction value of carbon emissions and the actual value of carbon emissions; configuring model parameters of the neural network model to make the loss function value lower than a preset threshold; determining configured neural network model as prediction model of the target type. Therefore, based on historical operation data of each type of device and actual value of carbon emissions of the type of device, it is convenient to train prediction model of carbon emissions for each type of device.
-
In one embodiment, wherein the operation data comprises a general data type and a special data type, the general data type is applicable to all types of devices in the industrial process, and the special data type is applicable to the target type. Therefore, operation data can be implemented in many ways, which not only reflects individual attributes of carbon emissions of specific device, but also reflects common attributes of carbon emissions between different devices.
-
In one embodiment, wherein obtaining operation data of a type of device during a historical execution period of the industrial process comprises: deploying a detachable electricity meter for the device of target type before the historical execution period of the industrial process; acquiring a reading of the electricity meter during the historical execution period of the industrial process; converting the reading into the actual value of carbon emissions; removing the detachable electricity meter from the device. Therefore, the actual value of carbon emissions of the type of device during the historical execution period can be easily determined through detachable electricity meter.
-
For example, suppose that the industrial process is a SMT (Surface Mount Technology) production line, and a device type is SMT machine. The process of determining the carbon emission prediction model of SMD (Surface Mount Device) machine includes the following steps: (1) : before running the SMD production line, arrange a detachable electricity meter for the SMD machine. (2) : acquire reading of the electricity meter in the historical execution period of the production line and operation data of a SMT machine in the historical execution period. For example, the operation data can include: device utilization rate and idle time as a general data type, and transit time and placement success rate as a special data type; (3) : convert the reading of the electricity meter into actual value of carbon emissions; (4) : remove the electricity meter; (5) : determine training data set, which includes the operation data and the actual value of carbon emissions as a label; (6) : input the training data set into an neural network model, wherein the neural network model is suitable for outputting a prediction value of carbon emissions based on the operation data; (7) : determine loss function value of the neural network model based on the difference between the prediction value of carbon emissions and the actual value of carbon emissions; (8) : configure model parameters of the neural network model to make the loss function value lower than a preset threshold. After the carbon emission prediction model of the SMD machine is generated, operation data during real-time execution period of the SMD production line can be input into the carbon emission prediction model to predict real-time carbon emission of SMD machines.
-
Specifically, the neural network model can be implemented as: feedforward neural network model, radial basis function neural network model, short-and long-term memory (LSTM) network model, echo state network (ESN) , gate loop unit (GRU) network model or deep residual network model, and so on.
-
The above exemplary description describes typical embodiments of the neural network model. Those skilled in the art can realize that this description is only exemplary and is not used to limit the scope of protection of the embodiments of the disclosure.
-
Step: 203: determining prediction value of carbon emissions of the industrial process based on the number of types, the number of devices of each type and the prediction values for respective types of devices in the industrial
process.
-
In one embodiment, determining prediction value of carbon emissions of the industrial process based on the number of types, the number of devices of each type and the prediction values for respective types of devices in the industrial process comprises: determining respective prediction values of carbon emissions of all devices of respective same types based on respective numbers of devices of respective same types and respective prediction values output from respective prediction models of carbon emissions associated with respective same types ; calculating sum of respective prediction values of carbon emissions of all devices of respective same types; determining the sum as the prediction value of carbon emissions of the industrial process.. Therefore, by classifying the devices in the industrial process according to respective device types, and then summing up based on the classification results, prediction value of carbon emissions of the entire industrial process can be easily determined.
-
In addition, embodiments of the present disclosure introduce AI capability into the process of carbon emissions prediction, reducing workload of carbon emissions prediction. In addition, devices with the same type share a carbon emission prediction model, there is no need to build special model for each device, reducing implementation difficulty of carbon emissions prediction.
-
For example, suppose that devices of an industrial process include two mounting machines (mounting machine A and mounting machine B) , reflow soldering device C and optical detection device D.
-
First of all, input the operation data of SMD A and SMD B collected during historical execution period of the industrial process into the carbon emission prediction model of SMD respectively to obtain carbon emissions prediction value S1 of SMD A and S2 of SMD B. Input the operation data of reflow soldering equipment C collected during the historical execution period of the industrial process into the carbon emission prediction model of reflow soldering equipment to obtain carbon emissions prediction value S3 of reflow soldering equipment C. Input the operation data of optical detection device D collected during the historical execution period of the industrial process into the carbon emission prediction model of optical detection device to obtain carbon emission prediction value S4 of optical detection device D. sum S1～S1, and the sum value is the predicted carbon emissions of the industrial process.
-
FIG. 5 is a schematic diagram of the carbon emissions prediction process of industrial process according to embodiments of the present disclosure.
-
As shown in FIG. 5, user 39 selects one or more devices in device-list area 31 of visual interface 30, and drags the selected one or more devices to target area 32 of visual interface 30. For example, user 39 drags out SMD machine 33, reflow soldering 34, buffer machine 35 and optical inspection equipment 36 to form a
mounting production line. The input parameter set 37 contains real-time operation data of SMD machine 33, reflow soldering 34, buffer machine 35 and optical inspection equipment 36 during the mounting production process. The real-time operation data of SMD machine 33 is provided to carbon emission prediction model of SMD machine. The real-time operation data of reflow soldering 34 is provided to carbon emission prediction model of reflow soldering. The real-time operation data of the cache machine 35 is provided to carbon emission prediction model of cache machine. The real-time operation data of the optical inspection equipment 36 is provided to carbon emission prediction model of the optical inspection equipment. Then, sum output value of carbon emission prediction model of SMD machine, output value of carbon emission prediction model of the reflow soldering, output value of carbon emission prediction model of cache machine and output value of the carbon emission prediction model of optical inspection equipment, and the sum value 38 is prediction value of carbon emissions of the mounting production line.
-
FIG. 6 is a structural diagram of apparatus for predicting carbon emissions from an industrial process according to embodiments of the present disclosure. The apparatus 500 comprising: a first determining module 501, configured to determine the number of types of devices comprised in an industrial process and the number of devices of each type; an obtaining module 502, configured to obtain prediction values of carbon emissions for respective types of devices in the industrial process, wherein the prediction values of carbon emissions for respective types of devices are output from respective prediction models of carbon emissions associated with respective types of devices, and each prediction model of carbon emissions is determined through AI and a second determining module 503, configured to determine prediction value of carbon emissions of the industrial process based on the number of types, the number of devices of each type and the prediction values for respective types of devices in the industrial process.
-
Embodiments of the present disclosure also provide an electronic device with processor-memory architecture. FIG. 7 is a structural diagram of an electronic device according to embodiments of the present disclosure.
-
As shown in FIG. 7, the electronic device 600 comprises a processor 601, a memory 602, and a computer program stored in the memory 602 and executable on the processor 601. When the computer program is executed by the processor 601, any one of the above-mentioned methods for predicting carbon emissions from an industrial process or a method for determining a prediction model of carbon emissions is implemented. The memory 602 may specifically be implemented as various storage medium such as Electrically Erasable Programmable Read-Only Memory (EEPROM) , Flash memory (Flash memory) , Programmable Program Read-Only Memory (PROM) . The processor 601 may be implemented to include one or more central processing units or one or more field programmable gate arrays, wherein the field programmable gate arrays integrate one or more central
processing unit cores. Specifically, the central processing unit or central processing unit core may be implemented as a CPU or MCU or DSP, and so on.
-
It should be noted that not all steps and modules in the above-mentioned processes and structural diagrams are necessary, and some steps or modules may be omitted according to actual needs. The execution logic of each step is not fixed and can be adjusted as needed. The division of each module is only to facilitate the description of the functional division used. In actual implementation, a module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be in the same device or in a different device.
-
The hardware modules in various embodiments may be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuits or logic devices (e.g., special-purpose processors, such as FPGAs or ASICs) to perform specific operations. Hardware modules may also include programmable logic devices or circuits temporarily configured by software (e.g., including general-purpose processors or other programmable processors) for performing operations. As for the specific mechanical method, or a dedicated permanent circuit, or a temporarily configured circuit (e.g., configured by software) to realize the hardware module, it can be decided according to cost and time considerations.
-
The above descriptions are merely preferred embodiments of the present disclosure, and are not intended to limit the protection scope of the present disclosure. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included within the protection scope of the present disclosure.