Fuel consumption prediction model training, fuel consumption prediction method, device and operating machinery

Abstract

The present invention provides a fuel consumption prediction model training, a fuel consumption prediction method, a device and an operating machine, wherein the model training method comprises: collecting a plurality of operating condition data that affect fuel consumption, each operating condition data refers to data that affect fuel consumption in a working cycle; extracting the time interval between the activated oil level and the predicted oil level from each operating condition data, and labeling each operating condition data based on the time interval and the preset time period to obtain each operating condition data label; training the initial model based on each operating condition data and each operating condition data label to obtain a fuel consumption prediction model. The fuel consumption prediction model trained by the present invention can predict in advance whether the current oil level of the operating machine can meet the operating requirements after the preset time period when the current oil level of the operating machine reaches the activated oil level, so that the operating personnel can have sufficient time to replenish the fuel and avoid the problem of delays in the construction period due to insufficient fuel.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a fuel consumption prediction model training, a fuel consumption prediction method, a device and an operating machine. Background Art

[0002] The engine is the core of the power system of the operating machinery. Its fuel usage analysis can help operators understand the operating status of the operating machinery in a timely manner so as to make good response decisions.

[0003] The existing technology usually obtains the measurement of the fuel level in the fuel tank and the engine fuel consumption in real time to determine the current fuel level in real time, and issues an early warning when the current fuel level is lower than the preset fuel level. However, for equipment users in the field of operating machinery, such as in the timed and quantitative tasks in the field of excavators, operators often do not have the energy to pay attention to the fuel usage. When receiving a fuel alarm, the operator cannot predict the remaining fuel usage time in advance, and cannot accurately schedule their work tasks, and often encounters situations where insufficient fuel affects the construction period. In addition, since most operating machinery has a harsh working environment, the rough road surface causes the fuel tank to shake greatly, and the real-time monitoring data is accompanied by a lot of noise, it brings certain difficulties to the real-time fuel consumption prediction, and often leads to a certain degree of misjudgment. Summary of the invention

[0004] The present invention provides a fuel consumption prediction model training, a fuel consumption prediction method, a device and an operating machine, which are used to solve the defects in the prior art that the remaining use time of fuel cannot be predicted in advance and errors exist in real-time monitoring of fuel consumption.

[0005] The present invention provides a fuel consumption prediction model training method, comprising:

[0006] Collect multiple working condition data that affect fuel consumption, each working condition data refers to the data that affects fuel consumption in a working cycle;

[0007] Extracting the time interval between the activated oil level and the predicted oil level from each operating condition data, and marking each operating condition data based on the time interval and the preset time period to obtain each operating condition data label, each operating condition data label is used to indicate whether the current oil level of the operating machine meets the operating requirements after the preset time period;

[0008] Based on the data of each operating condition and the labels of each operating condition data, the initial model is trained to obtain the fuel consumption prediction model.

[0009] According to a fuel consumption prediction model training method provided by the present invention, the marking of each operating condition data based on the time interval and the preset period includes:

[0010] When the time interval is greater than the preset period, the corresponding operating condition data label is that the remaining oil volume meets the operating requirements within the preset period;

[0011] When the time interval is less than or equal to the preset period, the corresponding operating condition data label is that the remaining oil volume does not meet the operating requirements within the preset period.

[0012] According to a fuel consumption prediction model training method provided by the present invention, before extracting the time interval between the activated fuel level and the predicted fuel level from each operating condition data, the method further includes:

[0013] Data preprocessing is performed on each operating condition data; the data preprocessing includes at least one of downsampling processing, noise reduction processing and deduplication processing.

[0014] According to a fuel consumption prediction model training method provided by the present invention, the operating condition data includes at least one of oil level data, oil temperature data, engine speed data, ambient temperature data, cooling water temperature data and gear data.

[0015] The present invention also provides a fuel consumption prediction method, comprising:

[0016] Determine the current working condition data of the machine to be predicted;

[0017] Inputting the current operating condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model;

[0018] The fuel consumption prediction model is trained based on the fuel consumption prediction model training method as described above.

[0019] According to a fuel consumption prediction method provided by the present invention, after obtaining the fuel consumption prediction result output by the fuel consumption prediction model, the method further includes:

[0020] When the fuel consumption prediction result is that after the preset period of time, the current oil level of the operating machine is less than or equal to the preset oil level, an early warning prompt is issued.

[0021] The present invention also provides a fuel consumption prediction model training device, comprising:

[0022] A collection unit, used to collect multiple working condition data affecting fuel consumption, each working condition data refers to data affecting fuel consumption in a working cycle;

[0023] A labeling unit is used to extract the time interval between the activated oil level and the predicted oil level from each working condition data, and label each working condition data based on the time interval and the preset time period to obtain each working condition data label, each working condition data label is used to indicate whether the current oil level of the operating machine meets the operating requirements after the preset time period;

[0024] The training unit is used to train the initial model based on various operating condition data and labels of the operating condition data to obtain a fuel consumption prediction model.

[0025] The present invention also provides a fuel consumption prediction device, comprising:

[0026] A determination unit, used to determine the current working condition data of the machine to be predicted;

[0027] A prediction unit, used for inputting the current working condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model;

[0028] The fuel consumption prediction model is trained based on the fuel consumption prediction model training method as described above.

[0029] The present invention also provides a working machine, comprising: the fuel consumption prediction device as described above.

[0030] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of any one of the above-mentioned fuel consumption prediction model training methods, and / or implements the steps of any one of the above-mentioned fuel consumption prediction methods.

[0031] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements the steps of any of the above-mentioned fuel consumption prediction model training methods and / or implements the steps of any of the above-mentioned fuel consumption prediction methods.

[0032] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of any of the above-mentioned fuel consumption prediction model training methods, and / or implements the steps of any of the above-mentioned fuel consumption prediction methods.

[0033] The fuel consumption prediction model training, fuel consumption prediction method, device and operating machinery provided by the present invention mark each operating condition data based on the time interval between the activation oil level and the predicted oil level and the preset time period, so that the fuel consumption prediction model trained based on each operating condition data and each operating condition data label can predict in advance whether the current oil level of the operating machinery can meet the operating requirements after the preset time period when the current oil level of the operating machinery reaches the activation oil level, so that the operating personnel can have sufficient time to replenish fuel and avoid the problem of construction delays due to insufficient fuel. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is one of the flow charts of the fuel consumption prediction model training method provided by the present invention;

[0036] Figure 2 This is the second flow chart of the fuel consumption prediction model training method provided by the present invention;

[0037] Figure 3 It is a flow chart of the fuel consumption prediction method provided by the present invention.

[0038] Figure 4 It is a structural schematic diagram of the fuel consumption prediction model training device provided by the present invention;

[0039] Figure 5 It is a structural schematic diagram of the fuel consumption prediction device provided by the present invention;

[0040] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The existing technology usually obtains the measurement of the fuel level in the fuel tank and the engine fuel consumption in real time to determine the current fuel level in real time, and issues an early warning when the current fuel level is lower than the preset fuel level. However, for equipment users in the field of operating machinery, such as in the timed and quantitative tasks in the field of excavators, operators often do not have the energy to pay attention to the fuel usage. When receiving a fuel alarm, the operator cannot predict the remaining fuel usage time in advance, and cannot accurately schedule their work tasks, and often encounters situations where insufficient fuel affects the construction period. In addition, since most operating machinery has a harsh working environment, the rough road surface causes the fuel tank to shake greatly, and the real-time monitoring data is accompanied by a lot of noise, it brings certain difficulties to the real-time fuel consumption prediction, and often leads to a certain degree of misjudgment.

[0043] In this regard, the present invention provides a method and system for training a fuel consumption prediction model, which is used to predict whether the oil level of an operating machine will meet the operating requirements in the future. This method can not only provide timely input of work task scheduling for operators and improve work efficiency, but also reduce the impact of data noise, improve the accuracy of intelligent diagnosis, and enhance the machine owner experience.

[0044] Figure 1 is a flow chart of the fuel consumption prediction model training method provided by the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0045] Step 110: Collect a plurality of operating condition data that affect fuel consumption, where each operating condition data refers to data that affects fuel consumption in a working cycle.

[0046] Here, the operating condition data refers to parameters that affect fuel consumption, such as fuel level, fuel temperature, engine speed, cooling water temperature, ambient temperature, gear position, etc. Among them, the operating condition data can be collected in real time through sensors.

[0047] It should be noted that each operating condition data refers to the data that affects fuel consumption in a working cycle, wherein the process of the oil level dropping from 100% and approaching 0% is a working cycle of the operating machine, and the data within this time period can be used as operating condition data.

[0048] Optionally, the working condition data is determined based on the following steps: collecting the initial working condition data of the operating machinery within a period of time, and the initial working condition data is the data that affects the fuel consumption. Since the operating machinery may have multiple working cycles within a period of time, it is necessary to locate each working cycle. The embodiment of the present invention can use the local minimum value calculation method to locate the lowest point of each working cycle. For each lowest point of the oil level, search forward along the time point to the highest point, and the data between the highest point and the lowest point is the working condition data corresponding to one working cycle of the operating machinery.

[0049] Optionally, considering the impact of other operating condition data of the operating machinery on the oil level, the mean of the oil level slope in the range from the highest oil level to the activated oil level, the mean of the oil level, engine speed, oil pressure, cooling water temperature, the mean and standard deviation of the ambient temperature can be extracted as operating condition data.

[0050] Step 120: extract the time interval between the activated oil level and the predicted oil level from each operating condition data, and mark each operating condition data based on the time interval and the preset time period to obtain each operating condition data label. Each operating condition data label is used to characterize whether the current oil level of the operating machinery meets the operating requirements after the preset time period.

[0051] Specifically, the predicted oil level refers to the oil level that reaches the warning line. If the current oil level of the operating machine is less than or equal to the predicted oil level, it means that the remaining oil cannot meet the operating requirements. If the oil level warning is issued when the current oil level is less than or equal to the predicted oil level, the operating personnel will not have enough reaction time to decide on the response strategy, such as not having enough time to refuel, which will cause the operating machine to stop due to lack of oil, and thus fail to complete the operating task.

[0052] Therefore, an embodiment of the present invention also sets an activation oil level, and the activation oil level is lower than the preset oil level. When the current oil level reaches the activation oil level, it indicates that the remaining oil amount of the operating machinery is at risk of being lower than the predicted oil level after a preset period of time, and this preset period of time is sufficient for the operating personnel to react and make decisions.

[0053] When the time interval between the activated oil level and the predicted oil level is greater than the preset period, it indicates that after the preset period, the current oil level of the operating machinery will be greater than the predicted oil level, which can meet the operating requirements and does not require additional refueling. At this time, the corresponding working condition data label can be marked as 0; when the time interval between the activated oil level and the predicted oil level is less than or equal to the preset period, it indicates that after the preset period, the current oil level of the operating machinery will be less than or equal to the predicted oil level, which cannot meet the operating requirements and requires refueling in advance. At this time, the corresponding working condition data label can be marked as 1. Among them, the activated oil level, the predicted oil level, and the preset period can be set according to actual conditions, and the embodiment of the present invention does not specifically limit this. For example, the activated oil level can be set to a 40% full oil level, and the predicted oil level can be set to a 20% full oil level.

[0054] Step 130: Based on the operating condition data and the operating condition data labels, the initial model is trained to obtain a fuel consumption prediction model.

[0055] Specifically, after obtaining each operating condition data and labeling each operating condition data, the initial model is trained based on each operating condition data and each operating condition data label until the convergence condition is reached to obtain the fuel consumption prediction model. The convergence condition may be that the number of training times reaches a preset number of times, or that the loss function is less than a threshold value, which is not specifically limited in the embodiment of the present invention.

[0056] It can be understood that in order to verify the effect of the fuel consumption prediction model, the operating condition data can be divided into a training set and a test set in a ratio of 7:3, and the LightGBM model can be used for oil level classification to obtain the classification accuracy, recall rate and ROC curve (receiver operating characteristic curve) to verify the effect of the fuel consumption prediction model.

[0057] Compared with the traditional method of determining the current oil level through sensor measurement, the embodiment of the present invention predicts fuel consumption based on the fuel consumption prediction model, which can reduce the misjudgment problem caused by inaccurate sensor measurement and drastic changes in the operating conditions of the operating machinery, and can promptly remind the operating personnel to replenish fuel, thereby reducing construction delays caused by insufficient fuel.

[0058] The fuel consumption prediction model training method provided in an embodiment of the present invention labels each operating condition data based on the time interval between the activation oil level and the predicted oil level and the preset time period, so that the fuel consumption prediction model trained based on each operating condition data and each operating condition data label can predict in advance whether the current oil level of the operating machinery can meet the operating requirements after the preset time period when the current oil level of the operating machinery reaches the activation oil level, so that the operating personnel can have sufficient time to replenish fuel and avoid the problem of construction delays due to insufficient fuel.

[0059] Based on the above embodiment, each operating condition data is marked based on the time interval and the preset period, including:

[0060] When the time interval is greater than the preset period, the corresponding working condition data label is that the remaining oil volume meets the operating requirements within the preset period;

[0061] When the time interval is less than or equal to the preset period, the corresponding operating condition data label is that the remaining oil volume does not meet the operating requirements within the preset period.

[0062] Specifically, if the current oil level of the operating machine is less than or equal to the predicted oil level, it means that the remaining oil cannot meet the operating requirements. If the oil level warning is issued when the current oil level is less than or equal to the predicted oil level, the operator will not have enough reaction time to decide on a response strategy, such as not having enough time to refuel, resulting in the operating machine shutting down due to lack of oil, and thus failing to complete the operating task.

[0063] Therefore, an embodiment of the present invention also sets an activation oil level, and the activation oil level is lower than the preset oil level. When the current oil level reaches the activation oil level, it indicates that the remaining oil amount of the operating machinery is at risk of being lower than the predicted oil level after a preset period of time, and this preset period of time is sufficient for the operating personnel to react and make decisions.

[0064] When the time interval between the activated oil level and the predicted oil level is greater than the preset period of time, it indicates that after the preset period of time, the current oil level of the operating machinery will be greater than the predicted oil level, which can meet the operating requirements and does not require additional refueling. At this time, the corresponding operating condition data label can be that the current oil level of the operating machinery is greater than the preset oil level, such as marked as 0; when the time interval between the activated oil level and the predicted oil level is less than or equal to the preset period of time, it indicates that after the preset period of time, the current oil level of the operating machinery will be less than or equal to the predicted oil level, which cannot meet the operating requirements and needs to be refueled in advance. At this time, the corresponding operating condition data label can be that the current oil level of the operating machinery is less than or equal to the preset oil level, such as marked as 1.

[0065] Based on any of the above embodiments, before extracting the time interval between the activated oil level and the predicted oil level from each operating condition data, the method further includes:

[0066] Data preprocessing is performed on the data of each working condition; the data preprocessing includes at least one of downsampling processing, noise reduction processing and deduplication processing.

[0067] Specifically, due to the uncertainty of the construction site and working environment of the operating machinery (such as excavators), the operating machinery itself and its attached sensors are susceptible to bumps and vibrations, especially the oil level sensor. In the actual collected data, there is a lot of volatile noise. If the noisy data is directly used to train the fuel consumption prediction model, the accuracy of the model will be reduced. In addition, due to the limitation of sensor sensitivity, the data cannot reflect the changes in the operating machinery working condition data in a timely manner, and the data has a certain redundancy. Based on the above two reasons, after acquiring each working condition data, the embodiment of the present invention downsamples each working condition data, such as downsampling to 2.5 minutes. Among them, the specific downsampling frequency can be iteratively selected according to the quality of the model results. The specific sampling frequency value is not limited here. For example, when the model effect is better, the corresponding downsampling frequency is lower.

[0068] In addition, when collecting working condition data through sensors, the sensors often fail to collect data, which results in the oil level information not being collected continuously as the working time changes. In addition, the vibration of the operating machinery also brings certain volatility to the oil level data. In view of this part of the noise, after acquiring each working condition data, the embodiment of the present invention performs noise reduction processing on each working condition data. For example, if the oil level fluctuation at adjacent time points in a working cycle is greater than a certain threshold, the data is considered to be a noise point, and the interpolation method is used to smooth the part of the data or directly delete the working cycle data.

[0069] Furthermore, when the operating machine is in a shutdown state, the data acquisition system is shut down, and there may be data gaps in the oil level signal sorted by time. The embodiment of the present invention can directly delete the time record of the shutdown period.

[0070] During the actual operation of the operating machinery, there may be a state where the machine is turned on but not working, resulting in parameters such as the oil level remaining unchanged. This situation is unpredictable. After obtaining the data of each operating condition, the embodiment of the present invention deduplicates the data of each operating condition, for example, directly deleting duplicate record data.

[0071] It can be seen that through the above data preprocessing method, the operating condition data of a complete working cycle of the operating machinery's fuel use can be obtained. The operating condition data may include the time point and the oil level, oil temperature, engine speed, cooling water temperature, gear position, etc. corresponding to each time point.

[0072] Based on any of the above embodiments, the operating condition data includes at least one of oil level data, oil temperature data, engine speed data, ambient temperature data, cooling water temperature data and gear position data.

[0073] Specifically, oil level data, oil temperature data, engine speed data, ambient temperature data, cooling water temperature data and gear data are factors that affect fuel consumption. Using them as operating condition data to train the initial model can enable the obtained oil temperature prediction model to accurately obtain oil temperature prediction results.

[0074] It is understandable that when collecting the above-mentioned operating condition data, the above-mentioned operating condition data can be collected according to a preset frequency, such as the preset frequency can be 1 minute, 2 minutes, 3 minutes, etc.

[0075] Based on any of the above embodiments, the present invention provides a fuel consumption prediction model training method, such as Figure 2 As shown, the method includes:

[0076] First, collect the working condition data of the operating machinery, specifically including: downsampling the oil level of the operating machinery, with a downsampling frequency of 2.5 minutes. At the same time, assuming that the data discontinuity is caused by shutdown, delete the missing part of the oil level. If the fluctuation of adjacent points of the oil level data exceeds 10%, delete the working cycle. If the oil level data remains unchanged, delete the duplicate points. When the minimum oil level of the working cycle is greater than the activation oil level, the working cycle is considered invalid, and the corresponding working condition data is deleted. The time point when a single working cycle of the operating machinery is lower than 20% is determined by the method of local minimum positioning. Repeat the above process to finally obtain the working condition data of multiple valid working cycles of the operating machinery.

[0077] For the working condition data of each working cycle, features are extracted and labels are manually annotated, and the initial model is trained based on the extracted features and labels to obtain the fuel consumption prediction model.

[0078] Based on any of the above embodiments, the present invention provides a fuel consumption prediction method, such as Figure 3 As shown, the method includes:

[0079] Step 310, determining the current working condition data of the machine to be predicted;

[0080] Step 320: input the current operating condition data into the fuel consumption prediction model to obtain the fuel consumption prediction result output by the fuel consumption prediction model;

[0081] The fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model described in any of the above embodiments.

[0082] Specifically, after the trained fuel consumption prediction model is obtained, the fuel consumption prediction model can be used to perform fuel consumption prediction to obtain a fuel consumption prediction result, wherein the fuel consumption prediction result can be whether the oil level of the operating machine is lower than a preset oil level after a preset period of time.

[0083] For example, the activation oil level can be set to 40% of the full oil level, the predicted oil level can be set to 20% of the full oil level, and the preset time period is 2 hours. When the operating machinery oil level reaches the activation oil level, the operating condition data corresponding to the working cycle closest to the current time point of the activation oil level is extracted as the current operating condition data through the method in the above-mentioned fuel consumption prediction model, and the features in the current operating condition data are extracted, and the features in the current operating condition data are input into the fuel consumption prediction model for prediction. If the prediction result is that the operating machinery oil level will be lower than 20% of the full oil level (i.e., the preset oil level) after 2 hours, it indicates that the operating requirements cannot be met and fuel needs to be replenished in advance; if the prediction result is that the operating machinery oil level will be greater than 20% of the full oil level (i.e., the preset oil level) after 2 hours, it indicates that the operating requirements can be met and no additional refueling is required.

[0084] Based on any of the above embodiments, after obtaining the fuel consumption prediction result output by the fuel consumption prediction model, the method further includes:

[0085] When the fuel consumption prediction result shows that the current oil level of the operating machinery is less than or equal to the preset oil level after a preset period of time, an early warning prompt is issued.

[0086] Specifically, when the fuel consumption prediction result shows that the current oil level of the operating machinery is less than or equal to the preset oil level after a preset period of time, it indicates that the operating requirements cannot be met and fuel needs to be replenished in advance. Therefore, an early warning can be issued to enable the operating personnel to replenish fuel in time to avoid delays in the construction period due to insufficient fuel.

[0087] The fuel consumption prediction model training device provided by the present invention is described below. The fuel consumption prediction model training device described below and the fuel consumption prediction model training method described above can be referenced to each other.

[0088] Based on any of the above embodiments, the present invention provides a fuel consumption prediction model training device, such as Figure 4 As shown, the device comprises:

[0089] The collecting unit 410 is used to collect a plurality of working condition data affecting fuel consumption, each working condition data refers to data affecting fuel consumption in a working cycle;

[0090] The labeling unit 420 is used to extract the time interval between the activated oil level and the predicted oil level from each working condition data, and label each working condition data based on the time interval and the preset time period to obtain each working condition data label, and each working condition data label is used to indicate whether the current oil level of the operating machine meets the operating requirements after the preset time period;

[0091] The training unit 430 is used to train the initial model based on various operating condition data and labels of various operating condition data to obtain a fuel consumption prediction model.

[0092] Based on any of the above embodiments, the labeling unit 420 includes:

[0093] A first labeling unit is used for, when the time interval is greater than the preset period, corresponding to the working condition data label that the remaining oil volume within the preset period meets the working requirements;

[0094] The second marking unit is used to, when the time interval is less than or equal to the preset period, correspond to the operating condition data label as the remaining oil volume within the preset period not meeting the operating requirements.

[0095] Based on any of the above embodiments, it also includes:

[0096] The data processing unit is used to perform data preprocessing on each operating condition data before extracting the time interval between the activated oil level and the predicted oil level from each operating condition data; the data preprocessing includes at least one of downsampling processing, noise reduction processing and deduplication processing.

[0097] Based on any of the above embodiments, the operating condition data includes at least one of oil level data, oil temperature data, engine speed data, ambient temperature data, cooling water temperature data and gear position data.

[0098] Based on any of the above embodiments, the present invention also provides a fuel consumption prediction device, such as Figure 5 As shown, the device comprises:

[0099] A determination unit 510, used to determine the current working condition data of the working machine to be predicted;

[0100] A prediction unit 520, configured to input the current operating condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model;

[0101] Wherein, the fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model described in any of the above embodiments.

[0102] Based on any of the above embodiments, it also includes:

[0103] The prompt unit is used to issue an early warning prompt after obtaining the fuel consumption prediction result output by the fuel consumption prediction model and when the fuel consumption prediction result shows that the current oil level of the operating machinery is less than or equal to the preset oil level after the preset time period.

[0104] Based on any of the above embodiments, the present invention further provides an operating machine, comprising: a fuel consumption prediction device as described in any of the above embodiments.

[0105] Specifically, since the fuel consumption prediction device is activated when the current oil level of the operating machinery reaches the activation oil level, it can predict in advance whether the current oil level of the operating machinery can meet the operating requirements after a preset period of time, so that the operating personnel can have sufficient time to replenish the fuel and avoid delays in the construction period due to insufficient fuel.

[0106] Figure 6 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the fuel consumption prediction model training method, which includes: collecting a plurality of working condition data affecting fuel consumption, each working condition data refers to data affecting fuel consumption in a working cycle; extracting the time interval between the activated oil level and the predicted oil level from each working condition data, and based on the time interval and the preset time period, marking each working condition data to obtain each working condition data label, each working condition data label is used to characterize whether the current oil level of the operating machine meets the operating requirements after the preset time period; based on each working condition data and each working condition data label, training the initial model to obtain the fuel consumption prediction model.

[0107] And / or, to execute a fuel consumption prediction method, the method comprising: determining the current operating condition data of the operating machinery to be predicted; inputting the current operating condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model; wherein the fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model as described above.

[0108] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0109] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the fuel consumption prediction model training method provided by the above methods, and the method includes: collecting multiple operating condition data that affect fuel consumption, each operating condition data refers to data that affects fuel consumption in a working cycle; extracting the time interval between the activated oil level and the predicted oil level from each operating condition data, and based on the time interval and the preset time period, marking each operating condition data to obtain each operating condition data label, each operating condition data label is used to characterize whether the current oil level of the operating machinery meets the operating requirements after the preset time period; based on each operating condition data and each operating condition data label, training the initial model to obtain a fuel consumption prediction model.

[0110] And / or, to execute a fuel consumption prediction method, the method comprising: determining the current operating condition data of the operating machinery to be predicted; inputting the current operating condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model; wherein the fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model as described above.

[0111] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned fuel consumption prediction model training methods, the methods comprising: collecting a plurality of operating condition data affecting fuel consumption, each operating condition data refers to data affecting fuel consumption in a working cycle; extracting the time interval between the activated oil level and the predicted oil level from each operating condition data, and labeling each operating condition data based on the time interval and a preset time period to obtain a label for each operating condition data, each operating condition data label being used to characterize whether the current oil level of the operating machinery meets the operating requirements after the preset time period; training the initial model based on each operating condition data and each operating condition data label to obtain a fuel consumption prediction model.

[0112] And / or, to execute a fuel consumption prediction method, the method comprising: determining the current operating condition data of the operating machinery to be predicted; inputting the current operating condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model; wherein the fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model as described above.

[0113] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0114] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fuel consumption prediction model training method, characterized in that: include: Collect multiple working condition data that affect fuel consumption, each working condition data refers to the data that affects fuel consumption in a working cycle; wherein, the process in which the oil level drops from 100% and approaches 0% is a working cycle of the operating machine; Extracting the time interval between the activated oil level and the predicted oil level from each operating condition data, and marking each operating condition data based on the time interval and the preset time period to obtain each operating condition data label, each operating condition data label is used to characterize whether the current oil level of the operating machine meets the operating requirements after the preset time period; the predicted oil level refers to the oil level that reaches the warning line, and the activated oil level is greater than the predicted oil level; Based on the data of each working condition and the data labels of each working condition, the initial model is trained to obtain a fuel consumption prediction model; The marking of each operating condition data based on the time interval and the preset period includes: When the time interval is greater than the preset period, the corresponding operating condition data label is that the remaining oil volume meets the operating requirements within the preset period; When the time interval is less than or equal to the preset period, the corresponding operating condition data label is that the remaining oil volume does not meet the operating requirements within the preset period.

2. The fuel consumption prediction model training method according to claim 1, characterized in that: Before extracting the time interval between the activation oil level and the predicted oil level from the data of each operating condition, it also includes: Data preprocessing is performed on each operating condition data; the data preprocessing includes at least one of downsampling processing, noise reduction processing and deduplication processing.

3. The fuel consumption prediction model training method according to claim 1, characterized in that: The operating condition data includes at least one of oil level data, oil temperature data, engine speed data, ambient temperature data, cooling water temperature data and gear position data.

4. A fuel consumption prediction method, characterized in that: include: Determine the current working condition data of the machine to be predicted; Inputting the current operating condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model; Wherein, the fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model according to any one of claims 1 to 3.

5. The fuel consumption prediction method according to claim 4, characterized in that: After obtaining the fuel consumption prediction result output by the fuel consumption prediction model, the method further includes: When the fuel consumption prediction result is that after the preset period of time, the current oil level of the operating machine is less than or equal to the preset oil level, an early warning prompt is issued.

6. A fuel consumption prediction model training device, characterized in that: include: The collection unit is used to collect a plurality of working condition data affecting fuel consumption, each working condition data refers to data affecting fuel consumption in a working cycle; wherein, the process of the oil level decreasing from 100% and tending to 0% is a working cycle of the operating machine; a labeling unit, used to extract the time interval between the activated oil level and the predicted oil level from each operating condition data, and label each operating condition data based on the time interval and the preset time period to obtain each operating condition data label, each operating condition data label is used to indicate whether the current oil level of the operating machine meets the operating requirements after the preset time period; the predicted oil level refers to the oil level that reaches the warning line, and the activated oil level is greater than the predicted oil level; A training unit, used for training the initial model based on various working condition data and labels of the various working condition data to obtain a fuel consumption prediction model; The marking of each operating condition data based on the time interval and the preset period includes: When the time interval is greater than the preset period, the corresponding operating condition data label is that the remaining oil volume meets the operating requirements within the preset period; When the time interval is less than or equal to the preset period, the corresponding operating condition data label is that the remaining oil volume does not meet the operating requirements within the preset period.

7. A fuel consumption prediction device, characterized in that: include: A determination unit, used to determine the current working condition data of the machine to be predicted; A prediction unit, used for inputting the current working condition data into a fuel consumption prediction model to obtain a fuel consumption prediction result output by the fuel consumption prediction model; Wherein, the fuel consumption prediction model is trained based on the training method of the fuel consumption prediction model according to any one of claims 1 to 3.

8. A working machine, characterized in that: include: The fuel consumption prediction device as claimed in claim 7.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the training method of the fuel consumption prediction model as described in any one of claims 1 to 3 are implemented, and / or, when the processor executes the program, the steps of the fuel consumption prediction method as described in claim 4 or 5 are implemented.

Images