Excavator oil consumption prediction optimization method and system based on cloud platform

Abstract

The invention belongs to the technical field of excavator control, and discloses an excavator oil consumption prediction optimization method and system based on a cloud platform, and the method comprises the steps: obtaining the operation data of an excavator, constructing an oil consumption prediction model, constructing a composite loss function, and training the oil consumption prediction model through the composite loss function; based on the engine universal curve, the target oil consumption constraint and the feasibility boundary of the engine rotating speed and power, an oil consumption adjustment strategy table is obtained; acquiring real-time operation data of the excavator, inputting the real-time operation data into the trained oil consumption prediction model, outputting an oil consumption prediction value and confidence at the next moment, and judging whether to trigger operation data optimization or not; and if so, calculating an adjustment coefficient in combination with the confidence coefficient, and obtaining an operation data optimization value. According to the method, the computing power advantage of the cloud platform and the real-time interaction capability of the whole vehicle end are utilized, accurate prediction, dynamic optimization and closed-loop control of oil consumption are achieved, the limitation of traditional static control is fundamentally broken through, and the fuel economy potential of equipment is released.

Description

Technical Field

[0001] This invention relates to the field of excavator control technology, and in particular to a cloud platform-based method and system for predicting and optimizing excavator fuel consumption. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In the field of engineering construction, excavators, as core construction equipment, are directly affected by their fuel economy, which is crucial for controlling construction costs and improving operational efficiency, making it a key issue of concern in the industry. With the rapid development of IoT, cloud computing, and artificial intelligence technologies, traditional equipment control modes that rely on fixed parameters are no longer sufficient to meet the demands of refined and efficient construction. Therefore, how to overcome fuel consumption control bottlenecks through technological innovation and achieve dynamic optimization and precise control has become an important direction for excavator technology upgrades.

[0004] The core technology solution of existing excavator control systems is a "parameter fixation + static matching" model: First, the operating parameters of the equipment are obtained through actual measurements by operators during the prototype stage. Based on these measured data, a fixed control curve is preset and fixed into the equipment control system. During actual operation, the system only matches the basic fuel consumption data with the operating status according to the preset curve, without the need for real-time adjustment of core parameters. At the same time, the existing technology has basic data interaction capabilities: the operating data of the entire vehicle is uploaded to the cloud platform through TBOX (Telematics Terminal Box) to achieve real-time data feedback; the cloud platform can also send preset target parameters to the entire vehicle to complete the secondary matching of vehicle parameters, but this matching process is still based on fixed logic and does not incorporate dynamic optimization based on real-time operating conditions.

[0005] In summary, the existing technologies have the following technical problems: First, existing technologies rely on preset fixed control curves and cannot adjust parameters according to dynamic factors such as operator habits, real-time load changes, and working condition fluctuations. This makes it difficult to adapt to complex and ever-changing construction scenarios, resulting in the fuel economy potential not being fully realized.

[0006] Secondly, existing technologies cannot predict fuel consumption trends in advance and can only operate passively based on historical fixed parameters, making it difficult to achieve the target fuel consumption control of "advance adjustment and precise achievement".

[0007] Furthermore, although existing technologies have enabled data interaction between the vehicle and the cloud platform, they have not leveraged the advantages of cloud computing power for in-depth time-series modeling and intelligent optimization. The data is only used for basic feedback rather than optimization decisions, resulting in low efficiency in fuel consumption control. Summary of the Invention

[0008] To address the aforementioned issues, this invention proposes a cloud platform-based method and system for predicting and optimizing excavator fuel consumption. This invention fully leverages the computing power advantages of the cloud platform and the real-time interaction capabilities of the vehicle to achieve accurate fuel consumption prediction, dynamic optimization, and closed-loop control, fundamentally breaking through the limitations of traditional static control and unleashing the fuel economy potential of the equipment.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a cloud platform-based method for predicting and optimizing excavator fuel consumption, comprising the following steps: Acquire excavator operating data and preprocess the operating data; A fuel consumption prediction model is constructed, taking preprocessed operating data as input and predicting fuel consumption as output. A composite loss function is constructed, which includes a basic loss term, an asymmetric penalty term, a time-period weighting term, a trend consistency term, and an L2 regularization term. The fuel consumption prediction model is trained using the composite loss function. Based on the predicted fuel consumption value, the engine universal curve, the target fuel consumption constraint, and the feasibility boundary of engine speed and power, the optimal combination of control parameters is obtained through the PSO algorithm. A fuel consumption adjustment strategy table is then established based on the optimal combination of control parameters. The system acquires real-time operating data of the excavator, inputs it into a trained fuel consumption prediction model, outputs the fuel consumption prediction value for the next moment, calculates the confidence level of the fuel consumption prediction value, and determines whether to trigger operating data optimization. If triggered, based on the real-time operating data, it extracts the optimal speed-power vector from the fuel consumption adjustment strategy table, calculates the adjustment coefficient in combination with the confidence level, and obtains the optimized operating data value.

[0010] As an optional implementation, the acquired excavator vehicle-side operating data includes working gear, main pump pressure, main pump solenoid valve current, hydraulic oil temperature, engine coolant temperature, actual engine speed, real-time fuel injection quantity, fuel flow rate, and engine load rate; the preprocessing is to filter out outliers in the excavator vehicle-side operating data.

[0011] As an alternative implementation, the composite loss function is defined as: ;in, , , , They are respectively , , , The weight, The regularization coefficient is . This is a regularization term.

[0012] As an alternative implementation, the Represented as: ;in, Using Huber loss, For parameters loss function, This is the actual fuel consumption value. To predict fuel consumption; The Represented as: ;in, For parameters Asymmetric loss function; The Represented as: ;in, For the first The weights of each sample, Load rate; The Represented as: .

[0013] As an alternative implementation method, the specific method for determining whether to trigger operational data optimization is as follows: when the predicted fuel consumption value is less than the target value and the confidence level is greater than 90%, operational data optimization is not triggered; when the predicted fuel consumption is greater than the target value or the confidence level is greater than 90%, operational data optimization is triggered.

[0014] As an alternative implementation method, the runtime data optimization method is as follows: ; in, This is the vector of power at the current speed. This is the vector of optimal speed power extracted from the fuel consumption adjustment strategy table. ;in, The adjustment factor related to the confidence level. , These are the minimum and maximum adjustment parameters, respectively. For the Sigmoid function, For confidence level, The confidence threshold. This is the steepness parameter.

[0015] Secondly, the present invention provides a cloud platform-based excavator fuel consumption prediction and optimization system, comprising the following modules: The data acquisition module is configured to acquire excavator operating data and preprocess the operating data. The prediction model building module is configured to: construct a fuel consumption prediction model, taking preprocessed operating data as input and predicted fuel consumption value as output; construct a composite loss function including a basic loss term, an asymmetric penalty term, a time-period weighting term, a trend consistency term, and an L2 regularization term; train the fuel consumption prediction model using the composite loss function; and obtain the optimal control parameter combination through the PSO algorithm based on the predicted fuel consumption value, the engine universal curve, the target fuel consumption constraint, and the feasibility boundary of engine speed and power, and establish a fuel consumption adjustment strategy table based on the optimal control parameter combination. The optimization module is configured to: acquire real-time operating data of the excavator, input it into the trained fuel consumption prediction model, output the fuel consumption prediction value for the next moment, calculate the confidence level of the fuel consumption prediction value, and determine whether to trigger operating data optimization; if triggered, based on the real-time operating data, extract the optimal speed power vector from the fuel consumption adjustment strategy table, calculate the adjustment coefficient in combination with the confidence level, and obtain the optimized operating data value.

[0016] Thirdly, the present invention provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, complete the excavator fuel consumption prediction and optimization method based on a cloud platform as described in the first aspect.

[0017] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the excavator fuel consumption prediction and optimization method based on a cloud platform as described in the first aspect.

[0018] Fifthly, the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the excavator fuel consumption prediction and optimization method based on a cloud platform as described in the first aspect.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention uses excavator operating data as input to comprehensively capture real-time changes in operating conditions, rather than relying on fixed parameters. Based on real-time operating data, it predicts fuel consumption for the next moment. If optimization is triggered, it combines the "fuel consumption adjustment strategy table" generated by the PSO algorithm with confidence level calculations to calculate adjustment coefficients and output optimized operating data values. This allows the invention to adjust in real time according to changing operating conditions, avoiding fuel waste caused by a disconnect between static curves and actual operating conditions, and fully releasing the fuel economy potential under different operating scenarios.

[0020] This invention utilizes the synergistic effect of a basic loss term, an asymmetric penalty term, a time-based weighting term, a trend consistency term, and an L2 regularization term to make the predicted value closer to actual fuel consumption. It predicts fuel consumption at the next moment based on real-time operating data and triggers control based on confidence levels, rather than passively correcting after fuel consumption exceeds the target. This invention can identify the risk of exceeding fuel consumption limits in advance and ensure the stable achievement of the target fuel consumption through predictive control, avoiding the fuel consumption runaway problem caused by the "lack of prediction and difficulty in intervention" of existing technologies.

[0021] This invention relies on the computing power of a cloud platform to support LSTM model training and PSO algorithm optimization, transforming the acquired operational data into decision-making resources such as "predictive model parameters" and "adjustment strategy tables," rather than simply storing data. This solves the problem of low data value utilization in existing technologies, improves the intelligence level of fuel consumption control, and enhances fuel consumption control efficiency.

[0022] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0023] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0024] Figure 1 This is a working block diagram of a cloud platform-based excavator fuel consumption prediction and optimization method provided in Embodiment 1 of the present invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] It should be noted that the following detailed description is exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0027] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0028] Example 1 like Figure 1 As shown, this embodiment provides a cloud platform-based method for predicting and optimizing excavator fuel consumption. The method uses the operator's operating habits as input features and fuel consumption as output to train a model using LSTM. The optimal fuel consumption is then matched using the PSO algorithm. Next, based on the vehicle's data received from the cloud platform over the past hour, the method predicts whether the fuel consumption will exceed the target. If it does, corresponding parameter adjustments are made and sent to the vehicle for execution. The specific steps include: S1: Obtain excavator operation data and preprocess the operation data.

[0029] In step S1, the excavator operation data is the data transmitted back from the excavator vehicle end. Feature extraction is performed using three representative vehicles of each tonnage. After multiple training applications, nine features were found to be strongly correlated with the fuel consumption prediction. Specifically, the obtained excavator vehicle end operation data includes working gear, main pump pressure, main pump solenoid valve current, hydraulic oil temperature, engine coolant temperature, actual engine speed, real-time fuel injection quantity, fuel flow rate, and engine load rate.

[0030] In step S1, after obtaining the excavator operation data, it is necessary to preprocess the excavator operation data. In this embodiment, the preprocessing is to filter out outliers in the excavator vehicle operation data to avoid invalid data interfering with subsequent modeling, and at the same time to provide a reliable data foundation for subsequent prediction and optimization.

[0031] S2: Construct a fuel consumption prediction model, taking preprocessed operating data as input and predicted fuel consumption value as output. Construct a composite loss function that includes a basic loss term, an asymmetric penalty term, a time period weighting term, a trend consistency term, and an L2 regularization term. Use the composite loss function to train the fuel consumption prediction model. Based on the predicted fuel consumption value, the engine universal curve, the target fuel consumption constraint, and the feasibility boundary of engine speed and power, obtain the optimal control parameter combination through the PSO algorithm, and establish a fuel consumption adjustment strategy table based on the optimal control parameter combination.

[0032] In step S2, the constructed fuel consumption prediction model is an LSTM network model. During training, a composite loss function is used, which includes a basic loss term, an asymmetric penalty term, a time-period weighting term, a trend consistency term, and an L2 regularization term. The composite loss function is defined as follows: ;in, , , , They are respectively , , , The weight, The regularization coefficient is . This is a regularization term. In this embodiment, Take 1, Take 0.8, Take 0.5, Take 0.1, Take 0.001.

[0033] , is represented as: ;in, For parameters loss function, This is the actual fuel consumption value. To predict fuel consumption; when the prediction error is less than The rate of increase is penalized by a squared penalty to encourage rapid convergence; when the prediction error is greater than [a certain value], [further action is taken]. The speed is increased by 100 liters per hour, with a linear penalty to prevent outliers from dominating training. Select 1-1.5L / h.

[0034] Represented as: ;in, For parameters The asymmetric loss function. Severe penalties are imposed for overestimating performance to prevent misjudgments that could lead to incorrect adjustments, power reductions, or other disruptions to operational efficiency. When the predicted value... >True value ,but When the true value >Predicted value Then the loss = .in Choose 2.5.

[0035] The penalty for prediction errors under load conditions is increased, fuel consumption is higher under heavy loads, and arbitrary adjustments to normal parameters will affect operating efficiency, specifically as follows: ; , For the first The weights of each sample, Let be the load factor, where This is the time-period weighting adjustment factor, set to 0.5. A coefficient that is too low will result in insufficient differentiation between high and low load conditions; a coefficient that is too high will cause the prediction of low-load conditions to be ignored. When under light load... The smaller the sample size, the less the impact of the prediction error; under heavy load, The error of this sample is relatively large, and its contribution to the total loss is significant.

[0036] Represented as: ,in, , The trend consistency term ensures that the trend of the predicted curve matches the actual curve. This avoids drastic fluctuations in the predicted curve, which would cause frequent parameter adjustments in the control system and affect operational stability. Operators prefer "smooth but slightly inaccurate" control to "frequent adjustments."

[0037] In step S2, the engine universal curve, target fuel consumption constraint x (L / h), and feasible boundaries for engine speed and power are input. The fuel consumption adjustment strategy table is obtained using the PSO algorithm. In the feasible boundaries, the engine speed is set to 1400rpm-2200rpm, and the power range is controlled according to the engine tonnage. - .

[0038] S3: Obtain real-time operating data of the excavator, input it into the trained fuel consumption prediction model, output the fuel consumption prediction value for the next moment, calculate the confidence level of the fuel consumption prediction value, and determine whether to trigger operating data optimization; if triggered, based on the real-time operating data, extract the optimal speed power vector from the fuel consumption adjustment strategy table, calculate the adjustment coefficient in combination with the confidence level, and obtain the optimized operating data value.

[0039] In step S3, the vehicle sends relevant feature data (excavator real-time operation data) to the TBOX in real time via the CAN bus. The TBOX then transmits the data back to the cloud platform via the 4G network in real time. The cloud platform uses the excavator's real-time operation data from the past hour to predict the fuel consumption for the next hour.

[0040] In step S3, the specific method for determining whether runtime data optimization is triggered is as follows: If the predicted fuel consumption is less than the target value and the confidence level is greater than 90%, no operational data optimization will be triggered, and the vehicle will continue to be monitored without any adjustments. If the predicted fuel consumption is greater than the target value or the confidence level is greater than 90%, operational data optimization will be triggered. If the confidence level is less than 70% and the predicted fuel consumption is greater than the target fuel consumption + 20%, the cloud platform will issue relevant prompts, and manual adjustments will be made to the relevant data. Otherwise, no adjustments will be made.

[0041] The data optimization method is as follows: ; in, This is the vector of power at the current speed. This is the vector of optimal speed power extracted from the fuel consumption adjustment strategy table.

[0042] ;in, The adjustment factor related to the confidence level, when When =0, no adjustment is made; when When =1, perform the maximum adjustment; when 0 < When <1, make gradual adjustments.

[0043] , These are the minimum and maximum adjustment parameters, respectively; For the Sigmoid function; Confidence level; In this embodiment, the confidence threshold is used. This indicates a moderate variation in confidence level between 70% and 80%. This is the steepness parameter; this calculation method makes the adjustment smooth and without abrupt changes.

[0044] S4: The cloud platform sends commands (optimized operating data values) to the vehicle's TBOX via the 4G network. The TBOX then sends instructions (optimized operating data values) to the vehicle controller and engine controller via the CAN bus to adjust parameters. After adjustment, the data continues to be uploaded to the cloud platform for continuous prediction, thus completing the entire workflow.

[0045] Example 2 This embodiment provides a cloud platform-based excavator fuel consumption prediction and optimization system, including the following modules: The data acquisition module is configured to acquire excavator operating data and preprocess the operating data. The prediction model building module is configured to: construct a fuel consumption prediction model, taking preprocessed operating data as input and fuel consumption as output, and construct a composite loss function including a basic loss term, an asymmetric penalty term, a time period weighting term, a trend consistency term, and an L2 regularization term; train the fuel consumption prediction model using the composite loss function; and obtain a fuel consumption adjustment strategy table based on the engine universal curve, target fuel consumption constraints, and the feasibility boundaries of engine speed and power using the PSO algorithm. The optimization module is configured to: acquire real-time operating data of the excavator, input it into the trained fuel consumption prediction model, output the fuel consumption prediction value for the next moment, calculate the confidence level of the fuel consumption prediction value, and determine whether to trigger operating data optimization; if triggered, based on the real-time operating data, extract the optimal speed power vector from the fuel consumption adjustment strategy table, calculate the adjustment coefficient in combination with the confidence level, and obtain the optimized operating data value.

[0046] It should be noted that the above modules correspond to the steps in Embodiment 1, and the examples and application scenarios implemented by the above modules and their corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should also be noted that the above modules can be executed in a computer system as part of the system.

[0047] In further embodiments, the following is also provided: An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0048] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0049] A computer-readable storage medium for storing computer instructions that, when executed by a processor, perform the method of Embodiment 1.

[0050] The method in Example 1 can be directly executed by a hardware processor, or it can be executed by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0051] A computer program product includes a computer program that, when executed by a processor, implements the method in Embodiment 1.

[0052] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0053] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0054] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0055] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0056] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A cloud-based method for predicting and optimizing excavator fuel consumption, characterized in that, Includes the following steps: Acquire excavator operating data and preprocess the operating data; A fuel consumption prediction model is constructed, taking preprocessed operating data as input and predicting fuel consumption as output. A composite loss function is constructed, which includes a basic loss term, an asymmetric penalty term, a time-period weighting term, a trend consistency term, and an L2 regularization term. The fuel consumption prediction model is trained using the composite loss function. Based on the predicted fuel consumption value, the engine universal curve, the target fuel consumption constraint, and the feasibility boundary of engine speed and power, the optimal combination of control parameters is obtained through the PSO algorithm. A fuel consumption adjustment strategy table is then established based on the optimal combination of control parameters. The system acquires real-time operating data of the excavator, inputs it into the trained fuel consumption prediction model, outputs the fuel consumption prediction value for the next moment, calculates the confidence level of the fuel consumption prediction value, and determines whether to trigger operating data optimization. If triggered, based on real-time operating data, the optimal speed-power vector is extracted from the fuel consumption adjustment strategy table, the adjustment coefficient is calculated by combining the confidence level, and the optimized value of the operating data is obtained.

2. The excavator fuel consumption prediction and optimization method based on a cloud platform as described in claim 1, characterized in that, The acquired excavator vehicle-side operating data includes working gear, main pump pressure, main pump solenoid valve current, hydraulic oil temperature, engine coolant temperature, actual engine speed, real-time fuel injection quantity, fuel flow rate, and engine load rate; the preprocessing involves filtering out outliers from the excavator vehicle-side operating data.

3. The excavator fuel consumption prediction and optimization method based on a cloud platform as described in claim 1, characterized in that, The composite loss function is defined as follows: ;in, , , , They are respectively , , , The weight, The regularization coefficient is . This is a regularization term.

4. The excavator fuel consumption prediction and optimization method based on a cloud platform as described in claim 5, characterized in that, The Represented as: ;in, Using Huber loss, For parameters loss function, This is the actual fuel consumption value. To predict fuel consumption; The Represented as: ;in, For parameters Asymmetric loss function; The Represented as: ;in, For the first The weights of each sample, Load rate; The Represented as: .

5. The excavator fuel consumption prediction and optimization method based on a cloud platform as described in claim 1, characterized in that, The specific method for determining whether to trigger operational data optimization is as follows: if the predicted fuel consumption value is less than the target value and the confidence level is greater than 90%, then operational data optimization is not triggered; if the predicted fuel consumption is greater than the target value or the confidence level is greater than 90%, then operational data optimization is triggered.

6. The excavator fuel consumption prediction and optimization method based on a cloud platform as described in claim 1, characterized in that, The method for optimizing the runtime data is as follows: ； in, This is the vector of power at the current speed. This is the vector of optimal speed power extracted from the fuel consumption adjustment strategy table. ;in, The adjustment factor related to the confidence level. , These are the minimum and maximum adjustment parameters, respectively. For the Sigmoid function, For confidence level, The confidence threshold. This is the steepness parameter.

7. A cloud-based excavator fuel consumption prediction and optimization system, characterized in that, include: The data acquisition module is configured to acquire excavator operating data and preprocess the operating data. The prediction model building module is configured to: construct a fuel consumption prediction model, taking preprocessed operating data as input and predicted fuel consumption value as output; construct a composite loss function including a basic loss term, an asymmetric penalty term, a time-period weighting term, a trend consistency term, and an L2 regularization term; train the fuel consumption prediction model using the composite loss function; and obtain the optimal control parameter combination through the PSO algorithm based on the predicted fuel consumption value, the engine universal curve, the target fuel consumption constraint, and the feasibility boundary of engine speed and power, and establish a fuel consumption adjustment strategy table based on the optimal control parameter combination. The optimization module is configured to: acquire real-time operating data of the excavator, input it into the trained fuel consumption prediction model, output the fuel consumption prediction value for the next moment, calculate the confidence level of the fuel consumption prediction value, and determine whether to trigger operating data optimization. If triggered, based on real-time operating data, the optimal speed-power vector is extracted from the fuel consumption adjustment strategy table, the adjustment coefficient is calculated by combining the confidence level, and the optimized value of the operating data is obtained.

8. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, complete the excavator fuel consumption prediction and optimization method based on any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, complete the excavator fuel consumption prediction and optimization method based on a cloud platform as described in any one of claims 1-6.

10. A computer program product, characterized in that, The invention includes a computer program that, when executed by a processor, implements the excavator fuel consumption prediction and optimization method based on a cloud platform as described in any one of claims 1-6.

Images