METHOD AND SYSTEM FOR PREDICTING THE AGING OF A HEATING COMPONENT IN A THRACTION BATTERY MATERIAL COOKING SYSTEM

Abstract

This disclosure relates to a method for predicting the aging of a heating component in a traction battery material curing system, comprising: the acquisition (S11) of data from a heating rod in each of the temperature zones of a furnace during each of predefined time sub-periods, the training (S12) of different preliminary prediction models according to the heating rod data, and the filtering of the different trained preliminary prediction models to obtain an optimal prediction model; the input (S13) of heating rod data into the optimal prediction model to obtain a temperature prediction result;the subtraction (S14) of the temperature prediction result from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and the determination (S15) of a heating rod power if the difference value is outside a predefined range of difference values; if the heating rod power is greater than a predefined maximum power threshold, the determination of heating rod aging.

Description

Title of the invention: METHOD AND SYSTEM FOR PREDICTING THE AGING OF A HEATING COMPONENT IN A THRACTION BATTERY MATERIAL COOKING SYSTEM technical field

[0001] The present application relates to the technical field of temperature control of a heating rod, for example, a method and a system for predicting the aging of a heating component in a traction battery material baking system. Background

[0002] Industrial furnaces are equipment constructed with refractory materials for calcining materials or baking products. Industrial furnaces are widely used in industries such as machinery, metallurgy, and oil and gas. The creation and development of industrial furnaces play a very important role in human progress.

[0003] In the prior art, the main components of conventional industrial furnaces include the furnace masonry, a furnace purge system, a furnace preheater, a furnace combustion device, etc. In the usual process used in conventional industrial furnaces, to accelerate the cooking speed of the products and shorten the cooking cycle, the operator increases or decreases the fuel supply or modifies the power output to control the temperature inside the furnace. However, due to the instability of human control, the occurrence of excessive or insufficient power regulation is inevitable during manual operation, resulting in an excessive or delayed temperature rise in the furnace and affecting the sintering of the products.

[0004] During the material heating process, specific temperatures are maintained in multiple temperature zones by changing the power of heating rods, and the operating conditions of the heating rods differ according to the different temperature zones. When a heating rod fails to maintain a defined temperature value, it must be replaced promptly. However, due to the large number of heating rods in the workshop, workers cannot quickly identify all the abnormal heating rods. Summary

[0005] The present application provides a method and a system for predicting the aging of a heating component in a traction battery material baking system in order to obtain an intelligent prediction of the degree of aging of a heating rod.

[0006] In a first aspect, an embodiment of the present application provides a method for predicting the aging of a heating component in a traction battery material curing system. The method comprises the steps described below.

[0007] Data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace are acquired during each predefined sub-period of time of predefined sub-periods of time, the heating rod data comprising heating rod temperature data and heating rod power data.

[0008] Multiple different preliminary prediction models are trained on the heating rod data during the predefined time sub-period, and the multiple different trained preliminary prediction models are filtered to obtain an optimal prediction model.

[0009] Data from the heating rod from a predefined pre-prediction time period are entered into the optimal prediction model to obtain a temperature prediction result.

[0010] The temperature prediction result is subtracted from a predefined temperature value so as to obtain a difference value between the temperature prediction result and the predefined temperature value, and to detect a heating rod power in the case where the difference value is outside a predefined range of difference values.

[0011] In the event that the power of the heating rod is greater than a predefined maximum power threshold, an aging of the heating rod is determined.

[0012] Optionally, the method further comprises the following steps: the power of the heating rod is regulated according to a predefined control rule, and a temperature of a temperature zone where the heating rod is located is detected after a predefined time in the case where the power of the heating rod is less than the predefined maximum power threshold; and an aging of the heating rod is determined in the case where the temperature of the temperature zone where the heating rod is located remains unchanged, or an increasing value of the temperature of the temperature zone where the heating rod is located is less than a predefined temperature threshold.

[0013] Optionally, the method further comprises the following steps: information indicating that the heating rod needs to be replaced is sent in the event that aging of the heating rod is determined; data from the heating rod is acquired again after detecting a replacement of the heating rod to obtain data from the heating rod after the replacement of the heating rod;The heating rod data before heating rod replacement and the heating rod data after heating rod replacement in the same temperature zones of the different furnaces are analyzed using a plurality of statistical modules to obtain a trend of change before heating rod replacement and a trend of change after heating rod replacement in the same temperature zones of the different furnaces, the statistical modules including a mean value, variance, skewness and kurtosis coefficient; and a heating rod condition is determined according to the trend of change before heating rod replacement and the trend of change after heating rod replacement.

[0014] Optionally, the determination of the state of the heating rod according to the trend of change before the replacement of the heating rod and the trend of change after the replacement of the heating rod includes the following steps: the trend of change before the replacement of the heating rod is compared with the trend of change after the replacement of the heating rod in the same temperature zone of the same furnace; and / or the trend of change before the replacement of the heating rod is compared with the trend of change after the replacement of the heating rod in the same temperature zones of the different furnaces, to obtain the value of change of data of the replacement of the heating rod;if it is determined that preventive maintenance and replacement are carried out on the heating rod in the case where the data change value of the replacement of the heating rod is within a predefined range of the data change of the replacement of the heating rod; it is determined that the performance of the heating rod is unstable in the case where the data change value of the replacement of the heating rod is not within a predefined range of the data change of the replacement of the heating rod, and a difference value between the power of the heating rod before replacement and the power of the heating rod after replacement is not within a predefined power difference range;and if it is determined that a heating rod maintenance time record is incorrect in the case where a heating rod power fluctuation prior to replacement is outside a range of power fluctuations; predefined, a power fluctuation of the heating rod after replacement is within the predefined power fluctuation range, after the heating rod has heated up after replacement for a predefined heating time period, and a temperature of the heating rod after replacement is maintained at a specific temperature and a temperature fluctuation of the heating rod after replacement is within a predefined temperature fluctuation range.

[0015] Optionally, training the plurality of different preliminary prediction models according to the heating rod data during each predefined time sub-period, and filtering the different trained preliminary prediction models to obtain the optimal prediction model, involves the following steps: a prediction model for the heating rod is constructed, and a number of training runs and a model filtering condition corresponding to each training run are determined, the prediction model for the heating rod comprising a plurality of different pre-selected preliminary prediction models;a set of data to be entered for training and a set of data to be entered for testing in each of the training series are determined according to the heating rod data in a predefined time sub-period corresponding to each of the training series;In each of the training runs, the prediction model for the heating rod is trained and tested on the training input set and the test input set to deliver a test prediction error, the plurality of different preliminary prediction models in the prediction model for the heating rod are filtered according to the delivered test prediction error and a model filtering condition corresponding to a current training run to obtain a prediction review model for the heating rod, and the prediction review model for the heating rod is used as the prediction model for the heating rod in a subsequent training run to perform the training and testing of the prediction model for the heating rod in the next training run until the training runs are completed;and the optimal prediction model is determined according to a test prediction error delivered by a final training series and a model filtering condition corresponding to the final training series.

[0016] Optionally, in the case where the training series comprises two series, the predefined time sub-period comprises a predefined time period before the heating rod maintenance and a predefined time period after the heating rod maintenance corresponding to a first training series, and a predefined period of time before the prediction of the heating rod corresponding to a second training run, and the model filtering conditions include a first series filtering condition corresponding to the first training run and a second series filtering condition corresponding to the second training run; the first series filtering condition includes filtering a test prediction error into a first predefined test error range in the first training run, and selecting a preliminary prediction model corresponding to the filtered test prediction error into the first predefined test error range in the first training run, from prediction models for the heating rod of the first training run, to form a prediction review model for the heating rod of the first training run;and the second series filtering condition includes filtering a test prediction error with a minimum value in the second training series, selecting a preliminary prediction model corresponding to the test prediction error filtered with the minimum value, in prediction models for the heating rod of the second training series, and using the preliminary prediction model corresponding to the test prediction error filtered with the minimum value as the optimal prediction model.

[0017] Optionally, the method further comprises the following steps: prompt information is displayed on a display interface and voice information is sent to a predefined communication device in the event that aging of the heating rod is determined.

[0018] In a second aspect, an embodiment of the present application further provides a system for predicting the aging of a heating component in a traction battery material curing system. The system comprises a data acquisition module, a model training module, a temperature prediction module, and an aging prediction module.

[0019] The data acquisition module is configured to acquire data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace during each predefined sub-time period of predefined sub-time periods, the heating rod data comprising heating rod temperature data and heating rod power data.

[0020] The model training module is configured to train multiple different preliminary prediction models based on the heating rod data during the predefined time sub-period, and to filter the multiple models from Preliminary predictions are trained on different models to obtain an optimal prediction model.

[0021] The temperature prediction module is configured to deliver heating rod data for a predefined pre-prediction time period in the optimal prediction model to obtain a temperature prediction result.

[0022] The aging prediction module is configured to subtract the temperature prediction result from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and detect heating rod power in the case where the difference value is outside a predefined range of difference values; and in the case where the heating rod power is greater than a predefined maximum power threshold, determine heating rod aging.

[0023] In a third aspect, an embodiment of the present application further provides a computer device. The computer device comprises memory, a processor, and a computer program stored in memory and executable by the processor. When executing the computer program, the processor carries out the steps in any of the preceding methods.

[0024] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps in any of the preceding processes.

[0025] In summary, the present application provides a method and a system for predicting the aging of a heating component in a traction battery material curing system. The method comprises the steps described below. Data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace are acquired during each predefined time sub-period of a plurality of predefined time sub-periods, the heating rod data comprising heating rod temperature data and heating rod power data; multiple different preliminary prediction models are trained on the heating rod data during the predefined time sub-periods, and an optimal prediction model is obtained;Data from the heating rod over a predefined pre-prediction time period are entered into the optimal prediction model to obtain a temperature prediction result; the temperature prediction result is subtracted from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and to detect the heating rod power in the case where the difference value exceeds a predefined range; and in the case where the heating rod power is; When the power exceeds a predefined maximum threshold, the aging of the heating rod is determined. This application not only enables a rapid and intelligent prediction of the degree of aging of a heating rod, but also significantly improves the accuracy and rationality of the aging prediction result. Brief description of the drawings

[0026] [Fig-1] The [Fig. 1] is a flowchart of a method for predicting the aging of a heating rod according to an embodiment of the present application;

[0027] [Fig.2] Fig.2 is a schematic diagram of a system for predicting the aging of a heating rod according to an embodiment of the present application; and

[0028] [Fig.3] The [Fig.3] is a diagram illustrating the internal structure of a computer device according to an embodiment of the present application. Detailed description of the implementation methods

[0029] The present application is illustrated below in conjunction with the drawings and embodiments. The embodiments described below form part of the embodiments of this application and are used merely to illustrate the application, but are not intended to limit the scope of this application.

[0030] Traction battery materials generally comprise positive electrode materials and negative electrode materials. A common positive electrode material used in commercially available automotive power batteries is the ternary positive electrode material (nickel-cobalt-manganese (NiCoMn, i.e., NCM)). NCM processing typically involves seven steps: dosing and mixing, single sintering, crushing, coating, double sintering, sieving and iron removal, and packaging. Sintering is one of the important steps in pyrometallurgical NCM production. Time, temperature, and other factors during sintering can affect product performance. The common traction battery material firing system in the industry is usually a roller furnace, and heating rods are used as heating components within the roller furnace.A heating section, a heat storage section, and a cooling section are established within the roller furnace, and the temperatures of these multiple sections are controlled by heating rods. Therefore, predicting the aging of the heating rods is an important part of the production process for traction battery materials.

[0031] In one embodiment, as shown in [Fig. 1], an embodiment of the present application provides a method for predicting the aging of a component heating in a traction battery material cooking system. The process includes the steps described below.

[0032] In SI 1, data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace are acquired during each predefined sub-period of predefined sub-periods of time, the heating rod data comprising heating rod temperature data and heating rod power data.

[0033] A furnace has multiple temperature zones. Therefore, in this embodiment, when acquiring heating rod data, temperature zone element numbers can be configured for multiple temperature zones of the furnace so that the position of the heating rod in each temperature zone can be distinguished by the zone element number of each temperature zone. Then, a heating rod maintenance record for the heating rod in each temperature zone is acquired using the temperature zone element number, and the heating rod temperature data and heating rod power data in the heating rod maintenance record are used as heating rod data. In this embodiment, the heating rod maintenance record has two functions.As for the first function, the heating rod maintenance record is used to acquire a training dataset for a predictive model for the heating rod; in the first series of training and testing, the training dataset and the test dataset for the predictive model for the heating rod are usually acquired from the heating rod data in a predefined time period before the heating rod maintenance and a predefined time period after the heating rod maintenance.As for the second function, the heating rod maintenance recording is used to stop the collection of temperature detection or temperature prediction data in a temperature zone where a heating rod is located during a heating rod maintenance period in order to avoid any interference on the accuracy of the training data for the prediction model.

[0034] In this embodiment, import data is configured based on the heating rod maintenance record and the temperature zone element number. The aging of a heating rod is predicted, temperature change data is acquired according to the temperature zone element number corresponding to the heating rod, and the training dataset for the prediction model for the heating rod and the temperature data in a time step are filtered according to the maintenance record. of heating rod. The time step includes the duration of the data used to enter the prediction model for a prediction before the time when the prediction is needed.

[0035] In S12, multiple different preliminary prediction models are trained based on the heating rod data during each predefined time sub-period, and an optimal prediction model is obtained by filtering the multiple different trained preliminary prediction models.

[0036] In the embodiment, the step in which multiple different preliminary prediction models are trained according to the heating rod data during each predefined time sub-period, and where an optimal prediction model is obtained by filtering the multiple different trained preliminary prediction models, comprises the steps described below.

[0037] A prediction model for the heating rod is constructed, and training series and a model filtering condition corresponding to each training series are determined, the prediction model for the heating rod comprising multiple different pre-selected preliminary prediction models, and the training series come in a sequential order.

[0038] A set of data to be entered for training and a set of data to be entered for testing in each of the training series are determined according to the heating rod data in a predefined time sub-period corresponding to each of the training series.

[0039] In each of the training series, the prediction model for the heating rod is trained and tested on the training input set and the test input set so as to deliver a test prediction error, the multiple different preliminary prediction models in the prediction model for the heating rod are filtered according to the delivered test prediction error and a model filtering condition corresponding to a current training series so as to obtain a prediction review model for the heating rod, and the prediction review model for the heating rod is used as the prediction model for the heating rod in the next training series to perform the training and testing of the prediction model for the heating rod in the next training series until all the training series are performed.

[0040] The optimal prediction model is determined according to a test prediction error delivered by a final training series and a model filtering condition corresponding to the final training series.

[0041] In one embodiment, where the training series comprises two series, the predefined time sub-period comprises a time period predefined before and after heating rod maintenance corresponding to a first training series and a predefined time period before prediction for the heating rod corresponding to a second training series, and the model filtering conditions include a first series filtering condition corresponding to the first training series and a second series filtering condition corresponding to the second training series.

[0042] The first series filtering condition includes filtering a test prediction error into a first predefined range of test errors in the first training series, and selecting a preliminary prediction model corresponding to the filtered test prediction error in the first predefined range of test errors in the first training series, from prediction models for the heating rod of the first training series to form a prediction review model for the heating rod of the first training series.

[0043] The second series filtering condition involves filtering a test prediction error with a minimum value in the second training series, selecting a preliminary prediction model corresponding to the test prediction error filtered with the minimum value, in prediction models for the heating rod of the second training series, and using the preliminary prediction model corresponding to the test prediction error with the minimum value as the optimal prediction model.

[0044] Optionally, to ensure prediction accuracy, the embodiment selects at least two training series. Those skilled in the art can define the number of training series according to the specific implementation situation and select corresponding predefined time sub-periods based on the training series to perform training in each series on datasets from different predefined time sub-periods. A predefined time sub-period may include heating rod data throughout the heating rod's lifecycle or heating rod data before and after maintenance, which is not limited in the embodiment of this application.

[0045] By way of illustration, each predefined time sub-period selected in the embodiment can be long enough to fully reflect the characteristics of the heating rod data during training. However, a time period that is too long can lengthen the model training time. Therefore, the embodiment allows the selection of heating rod data for the 30 days prior to heating rod maintenance as the dataset. The specific time selection can be defined depending on the specific implementation situation. The heating rod data here includes heating rod temperature data and heating rod power data, and other related heating rod data can also be used here as a training and testing dataset.

[0046] In the embodiment, the prediction model for the heating rod is trained and filtered according to series, and different training sets are selected according to the different training series, so that the optimal prediction model agreeing with the current temperature zone and the properties of the current heating rod can be selected from among many prediction models for the heating rod in a hierarchical and targeted manner; therefore, the accuracy of the prediction is improved, and a reliable basis is provided for decision-making regarding the aging of the heating rod.

[0047] A preliminary prediction model for the embodiment can be a long-term and short-term memory (LSTM) model, a recurrent neural network (RNN) model, a convolutional neural network (CNN) model, or other deep learning models. Training and testing are performed repeatedly on these different preliminary prediction models in order to filter out the prediction model with the most accurate prediction result, thereby effectively ensuring the reliability of the results.

[0048] The embodiment uses the selected input model data to train the prediction model for the heating rod, and determines whether the prediction model training is complete based on the accuracy of the training and test results. If the accuracy falls within a predefined error range, the model can be used for prediction.

[0049] In S13, heating rod data from a predefined pre-prediction time period are entered into the optimal prediction model to obtain a temperature prediction result.

[0050] A person skilled in the art can define the duration of the pre-prediction time period according to the actual situation, which is not limited in the embodiment of the present application. For example, if an early warning concerning a heating rod must be given 12 hours in advance to allow sufficient time for material preparation, the temperatures in the following 12 hours are predicted. Since heating power data is also relevant, the heating power data and the temperature data of the temperature zone during a previous time period are entered together as features to predict the temperature data of the heating rod during the following 12 hours.

[0051] In S14, the temperature prediction result, subtracted from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, is obtained, and a heating rod power is detected in the case where the difference value is outside a predefined range of difference values.

[0052] In S15, in the case where the power of the heating rod is greater than a predefined maximum power threshold, an aging of the heating rod is determined.

[0053] In the event that the power of the heating rod is less than the predefined maximum power threshold, the power of the heating rod is regulated according to a predefined regulation rule, a temperature of a temperature zone where the heating rod is located is detected after a predefined time, and in the event that the temperature of the temperature zone where the heating rod is located remains unchanged, or an increasing value of the temperature of the temperature zone where the heating rod is located is less than a predefined temperature threshold, an aging of the heating rod is determined.

[0054] For example, when the difference between the predicted temperature and the predefined temperature exceeds the predefined range, such as when the predefined temperature of the temperature zone is set to 750 degrees Celsius and the temperature prediction model predicts that the temperature in the temperature zone is about to drop to 700 degrees Celsius, the heating rod's power is detected. If the power is below the predefined maximum power threshold, for example, 20%, the heating rod's power is regulated according to the defined rule, for example, the power can be regulated to 40%, and the heating rod's temperature is recorded again; if the heating rod's temperature rises significantly and steadily, it is indicated that the heating rod is functioning normally.If it is detected that the power of the heating rod is higher than the predefined maximum power threshold, for example, if the power reaches 105%, it is indicated that the heating rod is abnormal and must be replaced.

[0055] In one embodiment, the process provided in the embodiment further includes the step described below.

[0056] After determining that the heating rod is aging, information indicating that the heating rod needs to be replaced is sent.

[0057] If aging of the heating rod is determined according to the preceding steps, an audible and visual alarm may be emitted, or a text message may be sent to the communication device of the person in charge, or others Methods can be used to notify the relevant personnel as soon as possible in order to replace the heating rod in a timely manner.

[0058] After detecting a replacement of the heating rod, data from the heating rod are acquired again so as to acquire the data from the heating rod after the replacement of the heating rod.

[0059] The heating rod data before the heating rod replacement and the heating rod data after the heating rod replacement in the same temperature zones of the different furnaces are analyzed using multiple statistical modules so as to obtain a trend of change before the heating rod replacement and a trend of change after the heating rod replacement in the same temperature zones of the different furnaces, the multiple statistical modules having a mean value, a variance, a skewness and a kurtosis coefficient.

[0060] A state of the heating rod is determined according to the trend of change before the replacement of the heating rod and the trend of change after the replacement of the heating rod.

[0061] In the embodiment, the step in which the state of the heating rod is determined according to the trend of change before the replacement of the heating rod and the trend of change after the replacement of the heating rod comprises the steps described below.

[0062] A trend of change before the replacement of the heating rod in the same temperature zone of the same furnace is compared with a trend of change after the replacement of the heating rod in the same temperature zone of the same furnace, and / or, the trend of change before the replacement of the heating rod in the same temperature zones of the different furnaces is compared with the trend of change after the replacement of the heating rod in the same temperature zones of the different furnaces so as to obtain a value of change of data of the replacement of the heating rod.

[0063] In the case where the data change value of the heating rod replacement is within a predefined range of the data change of the heating rod replacement, if it is determined that preventive maintenance and replacement are carried out on the heating rod.

[0064] In the event that the value of the data change of the replacement of the heating rod is not within a predefined range of the data change of the heating rod replacement, and a value of difference between the power of the heating rod before replacement and the power of the heating rod after replacement is not within a predefined power difference range, it is determined that the performance of the heating rod is unstable.

[0065] In the case where a power fluctuation of the heating rod before replacement is not within a predefined power fluctuation range, a power fluctuation of the heating rod after replacement is within the predefined power fluctuation range, and after the heating rod has heated up after replacement for a predefined heating time period, a temperature of the heating rod after replacement is maintained at a specific temperature and a temperature fluctuation of the heating rod after replacement is within a predefined temperature fluctuation range, if it is determined that a heating rod maintenance time record is incorrect.A person skilled in the art can define the predefined ranges involved in the embodiment according to the actual implementation situation, which is not specifically limited in the embodiment of this application.

[0066] Optionally, after replacing the heating rod, a temperature change trend analysis is performed on the data before and after the heating rod maintenance using multiple statistical modules in the embodiment. If it is determined that parameters of the heating rod change significantly according to the temperature change trend, this indicates a difference in the stability and other characteristics of the heating rod.If it is determined that the data does not change significantly after the heating rod has been replaced according to the temperature change trend, this indicates that the maintenance record is incorrect, or that the heating rod after replacement still has problems, or that the temperature detection device has problems, and corresponding alarm information is then generated to remind the inspection.

[0067] For example, in the embodiment, when comparing and analyzing the data before heating rod maintenance and the data after heating rod maintenance of temperature zone 4u of furnace 5 (5#4u), the data before heating rod maintenance and the data after heating rod maintenance of temperature zone 4u of furnace 7 (7#4u) and the data before heating rod maintenance and the data after heating rod maintenance of temperature zone 4u of furnace 8 (8#4u), it is found that parameters of the heating rod change significantly before and after heating rod maintenance.However, the similarity between these three datasets is that average temperature values ​​after replacement of the heating rod are essentially around 750 degrees Celsius; the difference is that the standard deviations of the three datasets are . different, and that there are also significant differences in the power of the heating rods. This means that there are differences in the stability of the heating rods after replacement. Here, 5#4u, 7#4u, and 8#4u can also be heating rods in the same temperature ranges of different furnaces.

[0068] In the embodiment, when comparing and analyzing the data before and after heating rod maintenance for temperature zone 4u of furnace 2 (2#4u), the data do not change significantly after replacement. However, it is necessary to determine whether a situation of preventive maintenance and replacement exists. In the embodiment, it is necessary to determine whether the maintenance time recording is incorrect based on the power fluctuation before and after heating rod maintenance. For example, the heating rod power in temperature zone 4u of furnace 6 (6#4u) fluctuated before maintenance.After maintenance, the actual power of the heating rod is almost constant, and the heating rod temperature eventually fluctuates around 748.73 degrees Celsius after a brief heating process with a standard deviation of 2.53 degrees Celsius. This pattern is exactly the opposite of the patterns in the other three data sets for furnace 5 (5#), furnace 7 (7#), and furnace 8 (8#). Therefore, it is necessary to determine whether the maintenance time record is incorrect.

[0069] In the event that aging of the heating rod is detected, prompt information is displayed on a display interface and voice information is sent to a predefined communication device.

[0070] A future trend graph of the heating rod temperatures can be viewed through the interface, and the monitoring result is updated every 30 minutes. The update frequency of the monitoring result can also be set according to the user's needs, for example, once per hour, once every 10 minutes, etc. After the user has sequentially selected and entered the production line and the main burner on the interface, the interface displays all the models saved under that burner. If a specific temperature zone is clicked on the left, the temperature prediction trend graph for that specific temperature zone can be displayed. If the heating rod temperature prediction result is within a normal range, no temperature anomalies are displayed in the upper right corner of the interface.If the prediction result shows a temperature anomaly, the temperature anomaly is pushed to the upper right corner of the interface, and the team will be notified to check it in a timely manner via DingTalk (trademark), text messaging (SMS), work applications (APPs), or other means. If the abnormal heating rod is... When replaced, the heating rod prediction function can be temporarily suspended to avoid repeated alarms due to data misalignment.

[0071] The present application provides a method for predicting the aging of a heating component in a traction battery material baking system, which enables intelligent prediction of the degree of aging of a heating rod, facilitates the maintenance of a stable temperature of the heating rod, improves product performance, and reduces process anomalies.

[0072] Based on the previous method for predicting the aging of a heating component in a traction battery material curing system, an embodiment of the present application further provides a system for predicting the aging of a heating component in a traction battery material curing system. As shown in [Fig. 2], the system comprises a data acquisition module 1, a model drive module 2, a temperature prediction module 3, and an aging prediction module 4.

[0073] The data acquisition module 1 is configured to acquire data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace during each predefined sub-period of time of a plurality of predefined sub-periods of time, the heating rod data comprising heating rod temperature data and heating rod power data.

[0074] The model training module 2 is configured to train multiple different preliminary prediction models according to the heating rod data during each predefined time sub-period, and to obtain an optimal prediction model by filtering the multiple different trained preliminary prediction models.

[0075] The temperature prediction module 3 is configured to deliver the heating rod data for a predefined pre-prediction time period in the optimal prediction model to obtain a temperature prediction result.

[0076] The aging prediction module 4 is configured to subtract the temperature prediction result from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and to detect heating rod power in the case where the difference value is outside a predefined range of difference values; and in the case where the heating rod power is greater than a predefined maximum power threshold, to determine heating rod aging.

[0077] The aging prediction module 4 is further configured to, in the event that the power of the heating rod is below the predefined maximum power threshold, regulate the power of the heating rod according to a regulation rule predefined, detect a temperature of a temperature zone of the furnace where the heating rod is located after a predefined time, and in the case where the temperature of the temperature zone of the furnace is below a predefined temperature threshold, determine an aging of the heating rod.

[0078] Regarding the limitation of the system for predicting the aging of a heating component in a traction battery material curing system, reference may be made to the limitation of the method for predicting the aging of a heating component in a traction battery material curing system, which is not repeated here. The multiple modules of the preceding system may be implemented wholly or partially by means of software, hardware, or a combination of software and hardware. Each module described above may be integrated into or independent of a processor in a computing device in hardware form, or stored in memory in a computing device in hardware form so that the processor can invoke and execute an operation corresponding to each of the modules described above.

[0079] Figure 3 shows the internal structure of a computing device in one embodiment. The computing device can be a terminal or a server. As shown in Figure 3, the computing device comprises a processor, memory, a network interface, a display, and an input device, all connected via a system bus. The processor of the computing device is configured to provide computing and control capabilities. The memory of the computing device comprises non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for executing the operating system and computer programs stored on the non-volatile storage media. The network interface of the computing device is configured to communicate with an external terminal via a network connection.The display of the computing device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computing device may be a touch layer covered on the display screen, a key, a trackball or a touchpad located on the casing of the computing device, or may be a keyboard, a touchpad, a mouse with an external connection or similar.

[0080] It should be understood by persons with ordinary expertise in the art that the structure illustrated in [Fig. 3] is only a schematic diagram of a part of the structure related to the present application solution and does not limit the computer device to which the present application solution is applied, and a device Computer systems may include more or fewer components than those illustrated, or the assembly of certain components, or the same arrangement of components.

[0081] In summary, the present application provides a method and a system for predicting the aging of a heating component in a traction battery material curing system. The method comprises the steps described below. Data from a heating rod in each temperature zone of multiple temperature zones of a furnace are acquired during each predefined time sub-period of predefined time sub-periods, the heating rod data comprising heating rod temperature data and heating rod power data; multiple different preliminary prediction models are trained on the heating rod data during the predefined time sub-periods, and an optimal prediction model is obtained;Data from the heating rod over a predefined pre-prediction period are entered into the optimal prediction model to obtain a temperature prediction result; the temperature prediction result is subtracted from a predefined temperature value to obtain a difference between the temperature prediction result and the predefined temperature value, and the heating rod power is detected if the difference value exceeds a predefined range; and if the heating rod power exceeds a predefined maximum power threshold, the aging of the heating rod is determined. This application not only enables a fast and intelligent prediction of the degree of aging of a heating rod, but also significantly improves the accuracy and rationality of the aging prediction result.

[0082] The multiple embodiments in the description are described progressively. Identical or similar parts in the multiple embodiments are referred to one another. Each embodiment emphasizes the differences compared to the other embodiments. Since the system embodiments are substantially similar to the process embodiments, the description is relatively simple. For related content, reference can be made to a partial description of the process embodiments. It should be noted that multiple technical features of the preceding embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the multiple technical features in the preceding embodiments are described.However, as long as the combinations of these technical characteristics are not contradictory, these combinations must be interpreted as falling within the scope of disclosure.

[0083] Legend of [Fig. 1]:

[0084] SI 1: Acquire data from a heating rod in each temperature zone of multiple temperature zones of a furnace during each predefined sub-period of time.

[0085] S12: Train multiple different preliminary prediction models according to the heating rod data during each predefined time sub-period, and filter the different trained preliminary prediction models to obtain an optimal prediction model.

[0086] S13: Input the heating rod data for a predefined pre-prediction time period into the optimal prediction model to obtain a temperature prediction result.

[0087] S14: Subtract the temperature prediction result from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and detect a heating rod power in the case where the difference value is outside a predefined range of difference values.

[0088] S15: Determine an aging of the heating rod in the case where the power of the heating rod is greater than a predefined maximum power threshold.

Claims

Demands

1. A method for predicting the aging of a heating component in a traction battery material curing system, characterized in that it comprises: acquiring (SU) data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace during each predefined sub-period of time, wherein the heating rod data includes heating rod temperature data and heating rod power data; training (S 12) a plurality of different preliminary prediction models according to the heating rod data during each predefined sub-period of time, and filtering the different trained preliminary prediction models to obtain an optimal prediction model;the input (S 13) of heating rod data during a predefined pre-prediction time period into the optimal prediction model to obtain a temperature prediction result; the subtraction (S 14) of the temperature prediction result from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and the detection of heating rod power in the case where the difference value is outside a predefined range of difference values; and the determination (S 15) of heating rod aging in the case where the heating rod power is greater than a predefined maximum power threshold.

2. A method for predicting the aging of the heating component in the traction battery material curing system according to claim 1, further comprising: regulating the power of the heating rod according to a predefined control rule, and detecting the temperature of a temperature zone where the heating rod is located after a predefined time, in the case where the power of the heating rod is below the predefined maximum power threshold; and the determination of aging of the heating rod in the case where the temperature of the temperature zone where the heating rod is located remains unchanged, or an increasing value of the temperature of the temperature zone where the heating rod is located is less than a predefined temperature threshold.

3. A method for predicting the aging of the heating component in the traction battery material baking system according to claim 1, further comprising: sending information indicating that the heating rod in a temperature zone of the furnace must be replaced in the event that aging of the heating rod in a temperature zone of the furnace is determined; reacquiring data from the heating rod after detecting a replacement of the heating rod, to obtain data from the heating rod after the replacement of the heating rod;the analysis of heating rod data before heating rod replacement and heating rod data after heating rod replacement in the furnace temperature zone, using a plurality of statistical modules, to obtain a trend of change before heating rod replacement and a trend of change after heating rod replacement in the furnace temperature zone, in which the statistical modules include a mean value, a variance, a skewness and a kurtosis coefficient; and the determination of a heating rod state according to the trend of change before heating rod replacement and the trend of change after heating rod replacement.

4. A method for predicting the aging of the heating element in the traction battery material curing system according to claim 3, wherein the determination of the condition of the heating element based on the change trend before replacement of the heating element and the change trend after replacement of the heating element comprises: comparing the change trend before replacement of the heating element with the change trend after replacement of the heating element in the temperature range

5. of the oven; to obtain the data change value of the heating rod replacement; the determination of whether or not to carry out preventive maintenance and replacement on the heating rod, in the case where the data change value of the heating rod replacement is within a predefined range of the data change of the heating rod replacement; the determination of instability in the performance of the heating rod in the case where the data change value of the heating rod replacement is not within a predefined range of data change of the heating rod replacement, and a difference value between the power of the heating rod before replacement and the power of the heating rod after replacement is not within a predefined power difference range; and the determination specifying whether a heating rod maintenance time record is incorrect in the case where a heating rod power fluctuation before replacement is outside a predefined power fluctuation range, a heating rod power fluctuation after replacement is within the predefined power fluctuation range, and after the heating rod has heated up after replacement for a predefined heating time period, a heating rod temperature after replacement is maintained at a specific temperature and a heating rod temperature fluctuation after replacement is within a predefined temperature fluctuation range. A method for predicting the aging of the heating component in the traction battery material curing system according to claim 1, wherein the training (S 12) of a plurality of different preliminary prediction models based on the heating rod data during each predefined time sub-period, and the filtering of the different trained preliminary prediction models to obtain the optimal prediction model, includes: constructing a prediction model for the heating rod, and determining a number of training runs and a model filtering condition corresponding to each training run. training, wherein the prediction model for the heating rod comprises a plurality of different preselected preliminary prediction models; the determination of a data set to be entered for training and a data set to be entered for testing in each of the training series according to the heating rod data in a predefined time sub-period corresponding to each of the training series;in each of the training runs, the training and testing of the prediction model for the heating rod on the training input set and the test input set to deliver a test prediction error, the filtering of the plurality of different preliminary prediction models into the prediction model for the heating rod according to the delivered test prediction error and a model filtering condition corresponding to a current training run to obtain a prediction review model for the heating rod, and the use of the prediction review model for the heating rod as the prediction model for the heating rod in a subsequent training run to perform the training and testing of the prediction model for the heating rod in the subsequent training run until the training runs are performed;and the determination of the optimal prediction model according to a test prediction error delivered by a final training series and a model filtering condition corresponding to the final training series.

6. A method for predicting the aging of the heating component in the traction battery material curing system according to claim 5, wherein, where the drive series comprises two series, the predefined time sub-period comprises a predefined time period before heating rod maintenance and a predefined time period after heating rod maintenance corresponding to a first drive series, and a predefined time period before heating rod prediction corresponding to a second drive series, and the model filtering conditions comprise a first filtering condition of series corresponding to the first training series and a second series filtering condition corresponding to the second training series; the first series filtering condition includes filtering a test prediction error into a first predefined range of test errors in the first training series, and selecting a preliminary prediction model corresponding to the filtered test prediction error into the first predefined range of test errors in the first training series, from prediction models for the heating rod of the first training series, to form a prediction filtering model for the heating rod of the first training series;and the second series filtering condition includes filtering a test prediction error with a minimum value in the second training series, selecting a preliminary prediction model corresponding to the test prediction error filtered with the minimum value, in prediction models for the heating rod of the second training series, and using the preliminary prediction model corresponding to the test prediction error filtered with the minimum value as the optimal prediction model.;

7. A method for predicting the aging of the heating component in the traction battery material baking system according to claim 1, further comprising: displaying prompt information on a display interface and sending voice information to a predefined communication device in the event that aging of the heating rod is determined.

8. A system for predicting the aging of a heating component in a traction battery material curing system, comprising: a data acquisition module (1) configured to acquire data from a heating rod in each temperature zone of a plurality of temperature zones of a furnace during each predefined time sub-period, wherein the heating rod data includes heating rod temperature data and heating rod power data a model training module (2) configured to train a plurality of different preliminary prediction models based on the heating rod data during each predefined time sub-period, and filter the different trained preliminary prediction models to obtain an optimal prediction model; a temperature prediction module (3) configured to deliver the heating rod data for a predefined pre-prediction time period in the optimal prediction model to obtain a temperature prediction result; and an aging prediction module (4) configured to subtract the temperature prediction result from a predefined temperature value to obtain a difference value between the temperature prediction result and the predefined temperature value, and detect heating rod power in the case where the difference value is outside a predefined range of difference values; and in the case where the heating rod power is greater than a predefined maximum power threshold, determine heating rod aging.

9. A computer device, comprising a memory, a processor and a computer program stored in the memory and executable by the processor, wherein the processor, when executing the computer program, carries out the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps of the process according to any one of claims 1 to 7.

Images