Fuel efficiency optimization method and system based on operating condition identification

Abstract

This invention relates to the field of construction machinery control technology, specifically to a fuel efficiency optimization method and system based on working condition identification. The method involves acquiring environmental images surrounding the construction machinery, generating a disparity map, and performing 3D reconstruction. Based on the reconstructed image and a pre-saved joint model of the construction machinery, the 3D coordinates and spatial attitude of the boom joint are obtained. Based on the 3D coordinates and spatial attitude of the boom joint, the state parameters of the construction machinery, and the reconstructed image, the current working condition of the construction machinery is classified, and the action state and slope correction angle under different working condition types are determined. Based on the working condition type, action state, and slope correction angle, and using a pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation are obtained. The current state parameters of the construction machinery are then adjusted to achieve fuel optimization.

Description

Technical Field

[0001] This invention relates to the field of engineering machinery control technology, specifically to a fuel efficiency optimization method and system based on operating condition identification. 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] During the operation of construction machinery, the driver's experience is relied upon to switch working modes according to the working conditions, and the construction machinery itself cannot perceive the dynamic changes of the construction scene in real time. The driver can also combine the judgment results sent by various sensors in the construction machinery with experience to issue control commands to the construction machinery, but the sensor capabilities are limited and prone to misjudgment. For example, the tilt sensor cannot distinguish between "flat ground" and "small-angle slope repair" (<5°), and the pressure sensor is prone to misjudging "unloading empty" as "excavation load", resulting in fuel waste.

[0004] Taking the excavation event as an example, it is necessary to determine whether the excavation is in progress based on the handle signal and the high-pressure sensor value. However, the excavation event can only be determined after the excavation has started, which results in a delay. Summary of the Invention

[0005] To address the technical problems mentioned above, this invention provides a fuel efficiency optimization method and system based on operating condition recognition. It provides depth information through visual recognition and combines it with existing sensors in construction machinery to improve control performance while reducing control latency.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] The first aspect of the present invention provides a fuel efficiency optimization method based on operating condition identification, comprising the following steps:

[0008] Acquire environmental images around the construction machinery, generate a disparity map, and perform 3D reconstruction; based on the reconstructed images and a pre-saved joint model of the construction machinery, obtain the 3D coordinates and spatial orientation of the boom joint;

[0009] Based on the three-dimensional coordinates and spatial posture of the boom joint, the state parameters of the construction machinery, and the reconstructed image, the current working condition of the construction machinery is classified, and the action state and slope repair angle under different working condition types are determined.

[0010] Based on the working condition type, operating status, and slope repair angle, and using a pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation are obtained. The current state parameters of the construction machinery are then adjusted to achieve fuel optimization.

[0011] Furthermore, based on the reconstructed image and the pre-saved engineering machinery joint model, the three-dimensional coordinates and spatial pose of the boom joint are obtained. Specifically, based on the key point coordinates in the image, the model point coordinates in the engineering machinery joint model, and the intrinsic parameter matrix of the image acquisition unit, the 6-DOF pose of the boom joint in space is obtained by solving, and the pose matrix R|t is output.

[0012] Furthermore, the pre-saved engineering machinery joint model has at least four core joints of the engineering machinery.

[0013] Further, the solution involves: using the core joints as control points, determining the coordinates of the control points in the world coordinate system; converting the 3D points into a weighted combination of control points; using the intrinsic parameter matrix of the image acquisition unit to determine the projection equation; and constructing a system of linear equations to obtain the pose calculation formula.

[0014] Furthermore, the current working conditions of the construction machinery are classified, and the action status and slope repair angle under different working conditions are determined; specifically, the classification results are obtained by combining a pre-trained working condition classification model with a set classification strategy.

[0015] Furthermore, the slope angle is specifically defined as: slope angle θ = arctan(Δz / Δx), where Δz is the height difference between the bucket and the machine body, and Δx is the horizontal projection distance.

[0016] Furthermore, the load strength coefficient K L The mechanical load strength, K, is used to determine the overall reaction action state. L =(Hydraulic pressure weight × current hydraulic pressure percentage) + (slope weight × slope influence factor) + (speed weight × action speed coefficient),

[0017] Furthermore, the fuel supply correction factor is as follows: Actual fuel supply = calibrated fuel supply × β × (1 + 0.02 × (current speed - economic speed)); where β is the fuel supply correction factor, and the economic speed is a set value that varies under different operating conditions.

[0018] Furthermore, torque compensation is specifically defined as: final compensation torque ΔT = (basic compensation value × load strength coefficient K) L ) + 0.1 × (actual torque - target torque).

[0019] A second aspect of the present invention provides a fuel efficiency optimization system based on operating condition identification, comprising:

[0020] The visual recognition module is configured to acquire images of the environment surrounding the construction machinery.

[0021] The image processing module is configured to process the obtained environmental image, generate a disparity map, and perform 3D reconstruction.

[0022] The 3D estimation module is configured to obtain the 3D coordinates and spatial orientation of the boom joint based on the reconstructed image and the pre-saved engineering machinery joint model.

[0023] The working condition classification module is configured to classify the current working condition of the construction machinery based on the three-dimensional coordinates and spatial posture of the boom joint, the state parameters of the construction machinery, and the reconstructed image, and determine the action state and slope repair angle under different working condition types.

[0024] The fuel efficiency optimization module is configured to: based on the working condition type, action status, and slope repair angle, and using a pre-built fuel mapping database, obtain the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation, and adjust the current state parameters of the construction machinery to achieve fuel optimization.

[0025] Compared with existing technologies, one or more of the above technical solutions have the following beneficial effects:

[0026] 1. By providing depth information through visual recognition, it addresses subtle changes that traditional sensors cannot distinguish. Combined with existing sensors, it performs multimodal redundancy verification to avoid false triggering. Real-time acquired image data can be used to model the sequential actions of construction machinery based on time relationships, enabling more accurate understanding of the machinery's operating conditions and status. This reduces reliance on high-precision industrial sensors (such as LiDAR), thereby improving the overall control effect of construction machinery and lowering control costs.

[0027] 2. The pre-built fuel mapping database, acting as a dynamic parameter optimization engine, digitizes engineers' experience. By pre-converting discrete experience into a continuous parameter space, it eliminates the ambiguity of manual operation and ensures optimal parameters for any combination of working conditions. The mapping database defines the parameter sequence of the complete action chain. For example, in the five stages of loading (digging → lifting → positioning → unloading → returning), the parameters of each stage transition smoothly, avoiding sudden changes that increase fuel consumption.

[0028] 3. The load strength coefficient K is calculated based on the optimal parameters and the current state of the construction machinery. L It is not a fixed value, but a dynamic value calculated in real time. Similarly, the fuel supply correction factor β not only involves table lookup but also includes a speed compensation term, making it a dynamic value as well. This enables global optimization of fuel efficiency at the system level. Attached Figure Description

[0029] 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.

[0030] Figure 1 This is a schematic diagram of the overall process of the fuel efficiency optimization method based on operating condition identification provided in one or more embodiments of the present invention;

[0031] Figure 2 This is a schematic diagram of the structure of a fuel efficiency optimization system based on operating condition identification provided in one or more embodiments of the present invention. Detailed Implementation

[0032] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0033] It should be noted that the following detailed descriptions are 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.

[0034] As described in the background section, existing technologies rely on human experience to determine construction scenarios and working conditions in order to control construction machinery. Some existing technologies that use sensors combined with related recognition algorithms to determine construction conditions are prone to misjudgment and delays.

[0035] Therefore, the fuel efficiency optimization method and system based on working condition identification given in the following embodiments can identify the type of construction working condition (shoveling, loading, leveling, slope repair) and the action status (excavation, material moving, unloading, return, etc.) in real time and accurately. By using a pre-established working condition-fuel economy mapping database and combining the identified construction working condition type, the control parameters are dynamically optimized, thereby shortening the control response delay and improving fuel efficiency.

[0036] Example 1:

[0037] like Figure 1 As shown, the fuel efficiency optimization method based on operating condition identification includes the following steps:

[0038] Acquire environmental images around the construction machinery, generate a disparity map, and perform 3D reconstruction; based on the reconstructed images and a pre-saved joint model of the construction machinery, obtain the 3D coordinates and spatial orientation of the boom joint;

[0039] Based on the three-dimensional coordinates and spatial posture of the boom joint, the state parameters of the construction machinery, and the reconstructed image, the current working condition of the construction machinery is classified, and the action state and slope repair angle under different working condition types are determined.

[0040] Based on the working condition type, operating status, and slope repair angle, and using a pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation are obtained. The current state parameters of the construction machinery are then adjusted to achieve fuel optimization.

[0041] The hardware configuration for this embodiment is as follows:

[0042] High baseline distance binocular camera (baseline distance 12cm, f=4mm lens), deployed on the shock-absorbing bracket on the top of the cab;

[0043] Embedded domain controller (computing power ≥ 10 TOPS, supports INT8 quantization inference).

[0044] The camera acquires and preprocesses images of the surrounding area of ​​the construction machinery. The preprocessed images are then used for stereo matching to generate a disparity map. Point cloud reconstruction and keypoint detection are then performed to obtain the 3D coordinates of the boom joints. PnP pose estimation is then used to obtain a 3D pose estimate. Based on the 3D pose estimate, the working conditions are classified and the action state is determined. Simultaneously, the slope trimming angle is calculated, and the recognition result is generated and output.

[0045] The reconstructed image is used for working condition identification using a spatiotemporal fusion algorithm. Specifically, based on the reconstructed image and the three-dimensional coordinates of the boom joint, a three-dimensional attitude estimate for the construction machinery is obtained, and further, working condition labels, action status, and slope repair angle are obtained.

[0046] 3D pose estimation is performed by solving the spatial pose of the boom and bucket joints using the PnP algorithm (input: 2D key points + camera calibration parameters). Specifically, the coordinates of key points in the 2D images captured by the binocular camera are combined with the coordinates of the 3D model points of the construction machinery (such as an excavator) and the camera intrinsic parameter matrix. The PnP solver then solves for the 6-DOF pose (3D position + 3D rotation) of the boom joint in space, and the output pose matrix R|t is obtained.

[0047] In this embodiment, the coordinates of 12 2D keypoints are obtained through HRNet keypoint detection. The pose matrix R / t is obtained through the EPnP solver, and the 6-DOF pose, spatial coordinates, attitude angles, and spatiotemporal features are further determined. The slope trimming angle θ=arctanΔz / Δx is calculated based on the spatial coordinates and attitude angles. Based on the spatiotemporal features, the working conditions are classified using ResNet-34, and the state is further determined.

[0048] The 3D pose estimation algorithm is optimized based on the existing algorithm. The specific optimizations are shown in Table 1.

[0049] Table 1 Optimization of 3D pose estimation

[0050]

[0051] The key point detection network uses the lightweight HRNet (parameter count < 1.2M), and the input image is 1920x1080. Through multi-resolution parallel processing, three types of features are obtained: 256x144, 128x72, and 64x36. The feature fusion outputs a heatmap, and the key points are determined based on the heatmap.

[0052] The closed-form solution algorithm for EPNP is as follows:

[0053] (1) Definition of control points: Let the coordinates of the four control points in the world coordinate system be:

[0054] ;

[0055] in, , , Control points j Three-dimensional coordinates in the world coordinate system.

[0056] (2) 3D point parameterization:

[0057] Any 3D point P w i Represented as a weighted combination of control points:

[0058] ;

[0059] α ij For the weight of the centroid coordinates, i For 3D point indices (i=1,2,...,n).

[0060] (3) Camera projection equation:

[0061] ;

[0062] in, λ i is the depth scale factor, representing the distance from a 3D point to the camera. u i 、v i These are the pixel coordinates on the 2D image plane, and K is the camera intrinsic parameter matrix;

[0063] K=| fx 0 u0 |

[0064] K = | 0 fyv0 |

[0065] K = |001|.

[0066] in, fx, fyThe focal length (in pixels) is the focal length along the x and y axes. u0 The horizontal coordinates (in pixels) of the main point. v0 The vertical coordinates (in pixels) of the master point. C cj For the first j The 3D coordinates of each control point in the camera coordinate system.

[0067] (4) Construct a system of linear equations:

[0068] eliminate λ i The constraint equations are obtained as follows:

[0069] ;

[0070] ;

[0071] The pose calculation formula is as follows:

[0072] ;

[0073] in,[ u, v [R|t] represents the 2D image coordinates, K is the camera intrinsic parameter matrix, [R|t] is the rotation matrix + translation vector to be determined, and [X, Y, Z] are the 3D model coordinates.

[0074] The work condition classification model is based on an improved ResNet-34, with the following architecture: input layer (800x600x3), convolutional layer (64 filters, 7x7, stride=2) + BN + ReLU, max pooling (3x3, stride=2), [residual block groups x 4] (each group doubles the number of channels: 64→128→256→512), global average pooling, fully connected layer (1024 units) + Dropout (0.5), and output layer (4 units + softmax). The advantages of the work condition classification model compared to ResNet-34 are shown in Table 2.

[0075] Table 2 Advantages of the working condition classification model compared to ResNet-34

[0076]

[0077] The working condition classification strategy is shown in Table 3.

[0078] Table 3 Working Condition Classification Strategy

[0079]

[0080] Training strategy: Positive samples consist of 80,000 labeled images (20,000 for each working condition), and negative samples consist of 20,000 images of interference scenes (rain / fog / dust / night). The enhancement methods are dust simulation and motion blur. Dust simulation is achieved by adding Gaussian noise and reducing contrast by 30%. Motion blur is achieved by using linear blur with random directions (kernel size 15px).

[0081] Loss function: Total loss = alpha * cross-entropy loss + beta * triplet loss; where alpha = 0.8, to ensure basic classification accuracy; beta = 0.2, to improve feature discrimination ability.

[0082] Cross-entropy loss = -Σ [true label * log(predicted probability)]

[0083] Triple loss = max(0, ||anchor sample features - positive sample features||) 2 - ||Anchor Sample Features- Negative Sample Features|| 2 + boundary value); where || ||2 represents the squared Euclidean distance of the feature vectors, and the boundary value (margin) is set to 1.0.

[0084] The working condition types of construction machinery are obtained through a working condition classification model, including dumping, loading, leveling, and slope repair.

[0085] (1) Sliding condition:

[0086] [Slinging] --> Excavation and loading: Bucket height < 1.2m & hydraulic pressure > 28MPa;

[0087] Excavation and loading --> Rotation and positioning: Bucket height > 2.0m & machine body rotation > 30°;

[0088] Rotary positioning --> Spreading and unloading: Bucket reaches the unloading area & opening >75%;

[0089] Discharge and unload --> Return to reset: Pressure drop <15MPa & opening <20%;

[0090] Return to reset --> Excavation and loading: Bucket height < 1.0m & pointing towards the stockpile.

[0091] (2) Loading conditions:

[0092] [Loading] --> Precision digging: Bucket height < 0.8m & pressure > 30MPa;

[0093] Precise digging --> Smooth lifting: Bucket height > 1.5m & vibration < 0.1g;

[0094] Smooth improvement --> Mining card positioning: Mining card outline detected (visual);

[0095] Truck positioning --> Precise unloading: Bucket inside the truck bed & height <1.0m;

[0096] Precise unloading --> Efficient return: Opening degree > 85% & pressure < 18MPa;

[0097] Efficient return --> Precise excavation: Bucket height < 0.7m & pointing towards the material pile.

[0098] A rectangular outline (length > 5m, height > 2m) and texture feature "wheel + truck bed" were detected: confirmed as a mining truck.

[0099] (3) Flat ground conditions:

[0100] [Flat Ground] --> Scraper Grounding: Bucket Inclination Angle < 5° & Pressure < 15MPa;

[0101] Scraper grounding --> forward leveling: speed 0.3-0.8m / s & height fluctuation <5cm;

[0102] Forward leveling --> Lifting and turning: Body rotation > 20° & height > 0.3m;

[0103] Lifting plate for steering --> Reversing and positioning: Rotation angle reaches target value ±5°;

[0104] Reversal positioning --> scraper grounding: bucket re-touches the ground;

[0105] Forward leveling --> Operation completed: No elevation change for 3 consecutive times.

[0106] (4) Slope repair conditions:

[0107] [Slope Repair] --> Slope Location: Detected slope & gradient >10°;

[0108] Slope positioning --> Cutting into the slope: Bucket height < 0.5m & pressure > 25MPa;

[0109] Cut into the slope --> Slope adjustment: Move along the slope and change the slope by <1° / s;

[0110] Slope adjustment --> Fine-tuning inspection: Moving speed <0.2m / s;

[0111] Fine-tuning inspection --> Acceptance completed: Slope error <1° for 5 seconds.

[0112] Slope angle calculation:

[0113] θ = arctan(Δz / Δx), where Δz is the height difference between the bucket and the machine body, and Δx is the horizontal projection distance.

[0114] Based on the obtained operating condition type and action status, optimized control parameters are obtained using a pre-determined fuel mapping database.

[0115] By conducting preliminary experiments, the corresponding fuel control parameters for different operating conditions and action states are determined, forming a fuel mapping database, as shown in Table 4.

[0116] Table 4 Fuel Mapping Database

[0117]

[0118] Load strength coefficient K L The mechanical load strength is used to determine the overall reaction action state.

[0119] K L =(Hydraulic pressure weight × Current hydraulic pressure percentage) + (Slope weight × Slope influence factor) + (Speed ​​weight × Action speed coefficient). The meanings and calculation examples of each parameter during the calculation are shown in Table 5.

[0120] Table 5. Meanings of relevant parameters and calculation examples for calculating the load strength coefficient.

[0121]

[0122] Fuel supply correction factor β: Actual fuel supply = rated fuel supply × β × (1 + 0.02 × (current speed - economic speed)).

[0123] The economic speed reference is a set value. In this embodiment, the settings are: 1750 rpm for dumping, 1800 rpm for loading, 1700 rpm for level ground, and 1900 rpm for repairing slopes.

[0124] Torque compensation ΔT: Final compensation torque = (basic compensation value × load strength coefficient) + 0.1 × (actual torque - target torque). The meanings of the relevant parameters are explained in Table 6.

[0125] Table 6 Explanation of the meaning of parameters for torque compensation ΔT

[0126]

[0127] The above method enables precise identification of construction conditions (shoveling / loading / leveling / slope repair) and action states (digging / unloading, etc.) using only a camera. Combined with binocular vision 3D reconstruction and spatiotemporal fusion algorithms, the action state recognition latency is less than 200ms. Visual recognition can more accurately distinguish construction conditions, offering advantages in real-time performance and greater accuracy compared to traditional sensor data-based judgments. High-performance hardware ensures rapid computation of the visual processing algorithm, sending the results in real-time at 10ms intervals. Regarding the algorithm, traditional algorithms rely on vehicle sensor data. For example, in digging, traditional solutions require lever signals and high-pressure sensor values ​​to determine if digging has begun, resulting in latency. However, the visual algorithm, powered by high-performance hardware, can identify the digging process the instant it starts and send the results. This allows for different strategies to be implemented for different conditions, enabling more refined control strategies. Combined with a pre-determined fuel mapping database, it transforms engineers' discrete experience into a reusable continuous parameter space, improving control effectiveness while reducing latency.

[0128] In traditional construction machinery control, human experience has limitations. It is influenced by the driver's subjectivity and there is a delay between human observation, decision-making, and operation. Furthermore, the sensors within the construction machinery themselves have inherent shortcomings (such as misjudgment) and struggle to identify complex actions.

[0129] This solution provides depth information through visual recognition, addressing the issue of micro-slopes (such as 3° slope repairs) that tilt sensors cannot detect.

[0130] During visual recognition, multimodal redundancy verification is performed in conjunction with existing sensors. For example, during the "loading" operation, the outline of the mining truck (visual) and the sudden drop in unloading pressure (sensor) need to be detected simultaneously to avoid false triggering.

[0131] Real-time acquired image data can be used to model the sequential actions of construction machinery based on time relationships. For example, by analyzing the bucket trajectory of consecutive frames, it is possible to distinguish between "swinging" (fan-shaped motion) and "flat ground" (linear motion).

[0132] By combining visual recognition with existing sensors, the operating conditions and status of construction machinery can be more accurately grasped, reducing reliance on high-precision industrial sensors (such as LiDAR). Since the cost of binocular cameras is lower than that of high-precision industrial sensors, and the algorithm is transferable, it can be quickly adapted to equipment such as excavators and loaders by updating the definition of key points in the 3D model, thereby reducing the cost of the algorithm, improving the overall control effect of construction machinery and reducing control costs. Furthermore, the data continuously collected during the operation of construction machinery can help the algorithm to be continuously optimized.

[0133] A pre-built fuel mapping database, acting as a dynamic parameter optimization engine, digitizes engineers' experience. By pre-converting discrete experience into a continuous parameter space, it eliminates the ambiguity of manual operation and ensures optimal parameters for any combination of working conditions. The mapping database defines the parameter sequence of the complete action chain. For example, in the five stages of loading (digging → lifting → positioning → unloading → returning), the parameters transition smoothly at each stage, avoiding sudden changes that increase fuel consumption.

[0134] The load strength coefficient K is calculated based on the optimal parameters and the current state of the engineering machinery. L It is not a fixed value, but a dynamic value calculated in real time. Similarly, the fuel supply correction factor β not only involves table lookup but also includes a speed compensation term, making it a dynamic value as well. This enables global optimization of fuel efficiency at the system level.

[0135] Example 2:

[0136] A fuel efficiency optimization system based on operating condition identification includes:

[0137] The visual recognition module is configured to acquire images of the environment surrounding the construction machinery.

[0138] The image processing module is configured to process the obtained environmental image, generate a disparity map, and perform 3D reconstruction.

[0139] The 3D estimation module is configured to obtain the 3D coordinates and spatial orientation of the boom joint based on the reconstructed image and the pre-saved engineering machinery joint model.

[0140] The working condition classification module is configured to classify the current working condition of the construction machinery based on the three-dimensional coordinates and spatial posture of the boom joint, the state parameters of the construction machinery, and the reconstructed image, and determine the action state and slope repair angle under different working condition types.

[0141] The fuel efficiency optimization module is configured to: based on the working condition type, action status, and slope repair angle, and using a pre-built fuel mapping database, obtain the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation, and adjust the current state parameters of the construction machinery to achieve fuel optimization.

[0142] By providing depth information through visual recognition, subtle changes that traditional sensors cannot distinguish are addressed. Combined with existing sensors, multimodal redundancy verification is performed to avoid false triggering. Real-time acquired image data can be used to model the temporal movements of construction machinery based on time relationships, enabling more accurate understanding of the machinery's operating conditions and status. This reduces reliance on high-precision industrial sensors (such as LiDAR), thereby improving the overall control effect of construction machinery and lowering control costs.

[0143] A pre-built fuel mapping database, acting as a dynamic parameter optimization engine, digitizes engineers' experience. By pre-converting discrete experience into a continuous parameter space, it eliminates the ambiguity of manual operation and ensures optimal parameters for any combination of working conditions. The mapping database defines the parameter sequence of the complete action chain. For example, in the five stages of loading (digging → lifting → positioning → unloading → returning), the parameters transition smoothly at each stage, avoiding sudden changes that increase fuel consumption.

[0144] The load strength coefficient K is calculated based on the optimal parameters and the current state of the engineering machinery. L It is not a fixed value, but a dynamic value calculated in real time. Similarly, the fuel supply correction factor β not only involves table lookup but also includes a speed compensation term, making it a dynamic value as well. This enables global optimization of fuel efficiency at the system level.

[0145] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A fuel efficiency optimization method based on operating condition identification, characterized in that, Includes the following steps: Acquire environmental images around the construction machinery, generate a disparity map, and perform 3D reconstruction; based on the reconstructed images and a pre-saved joint model of the construction machinery, obtain the 3D coordinates and spatial orientation of the boom joint; Based on the three-dimensional coordinates and spatial posture of the boom joint, the state parameters of the construction machinery, and the reconstructed image, the current working condition of the construction machinery is classified, and the action state and slope repair angle under different working condition types are determined. Based on the working condition type, operating status, and slope repair angle, and using a pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation are obtained. The current state parameters of the construction machinery are then adjusted to achieve fuel optimization.

2. The fuel efficiency optimization method based on operating condition identification as described in claim 1, characterized in that, Based on the reconstructed image and the pre-saved engineering machinery joint model, the three-dimensional coordinates and spatial pose of the boom joint are obtained. Specifically, based on the coordinates of key points in the image, the coordinates of model points in the engineering machinery joint model, and the intrinsic parameter matrix of the image acquisition unit, the 6-DOF pose of the boom joint in space is obtained by solving, and the pose matrix R|t is output.

3. The fuel efficiency optimization method based on operating condition identification as described in claim 2, characterized in that, A pre-saved model of a construction machinery joint, containing at least four core joints of the construction machinery.

4. The fuel efficiency optimization method based on operating condition identification as described in claim 2, characterized in that, The solution involves: using the core joints as control points, determining the coordinates of the control points in the world coordinate system; converting the 3D points into a weighted combination of control points; using the intrinsic parameter matrix of the image acquisition unit to determine the projection equation; and constructing a system of linear equations to obtain the pose calculation formula.

5. The fuel efficiency optimization method based on operating condition identification as described in claim 1, characterized in that, The current working conditions of construction machinery are classified, and the action status and slope repair angle under different working conditions are determined. Specifically, the classification results are obtained by combining a pre-trained working condition classification model with a set classification strategy.

6. The fuel efficiency optimization method based on operating condition identification as described in claim 1, characterized in that, The slope angle is specifically defined as: slope angle θ = arctan(Δz / Δx), where Δz is the height difference between the bucket and the machine body, and Δx is the horizontal projection distance.

7. The fuel efficiency optimization method based on operating condition identification as described in claim 1, characterized in that, Load strength coefficient K L The mechanical load strength, K, is used to determine the overall reaction action state. L =(Hydraulic pressure weight × current hydraulic pressure percentage) +(slope weight ×slope influence factor) +(speed weight × action speed coefficient).

8. The fuel efficiency optimization method based on operating condition identification as described in claim 1, characterized in that, The fuel supply correction factor is as follows: Actual fuel supply = calibrated fuel supply × β × (1 + 0.02 × (current speed - economic speed)); where β is the fuel supply correction factor, and the economic speed is a set value that varies under different operating conditions.

9. The fuel efficiency optimization method based on operating condition identification as described in claim 1, characterized in that, Torque compensation, specifically: final compensated torque ΔT = (basic compensation value × load strength coefficient K) L ) + 0.1 × (actual torque - target torque).

10. A system for implementing the fuel efficiency optimization method based on operating condition identification as described in any one of claims 1-9, characterized in that, include: The visual recognition module is configured to acquire images of the environment surrounding the construction machinery. The image processing module is configured to process the obtained environmental image, generate a disparity map, and perform 3D reconstruction. The 3D estimation module is configured to obtain the 3D coordinates and spatial orientation of the boom joint based on the reconstructed image and the pre-saved engineering machinery joint model. The working condition classification module is configured to classify the current working condition of the construction machinery based on the three-dimensional coordinates and spatial posture of the boom joint, the state parameters of the construction machinery, and the reconstructed image, and determine the action state and slope repair angle under different working condition types. The fuel efficiency optimization module is configured to: based on the working condition type, action status, and slope angle, and using a pre-built fuel mapping database, obtain the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction coefficient, and torque compensation, and adjust the current state parameters of the construction machinery to achieve fuel optimization.

Images