Fuel efficiency optimization method and system based on working condition recognition

Abstract

The invention relates to the technical field of engineering machinery control, in particular to a fuel efficiency optimization method and system based on working condition recognition, and the method comprises the steps: obtaining an environment image around engineering machinery, generating a disparity map, and carrying out the three-dimensional reconstruction; according to the reconstructed image and a pre-stored engineering machinery joint model, obtaining three-dimensional coordinates and spatial postures of the movable arm joint; classifying the current working conditions of the engineering machinery according to the three-dimensional coordinates and the spatial attitude of the movable arm joint, the state parameters of the engineering machinery and the reconstructed image, and determining the action states and slope repairing angles under different working condition types; and according to the working condition type, the action state and the slope repairing angle, based on a pre-constructed fuel mapping database, obtaining a corresponding load intensity coefficient, an engine rotating speed, a main pump pressure, a fuel supply correction coefficient and torque compensation, and adjusting the current state parameters of the engineering machinery to realize fuel optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering machinery control, and in particular to a fuel efficiency optimization method and system based on working condition identification. Background Art

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

[0003] During operation, construction machinery relies on the driver's experience to switch operating modes based on working conditions. The machinery itself is unable to perceive dynamic changes in the construction scene in real time. Drivers can also combine judgment results from various sensors on the machinery with their experience to issue control commands. However, sensors have limited capabilities and are prone to misjudgment. For example, inclination sensors cannot distinguish between "flat ground" and "slightly angled slope repair" (less than 5°), and pressure sensors can easily misjudge "unloaded" as "excavated," resulting in fuel waste.

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

[0005] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a fuel efficiency optimization method and system based on working condition recognition, which provides depth information through visual recognition and combines with the existing sensors of engineering machinery to improve the control effect while reducing control delay.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A first aspect of the present invention provides a fuel efficiency optimization method based on operating condition identification, comprising the following steps: Acquire the environmental image around the construction machinery, generate a disparity map, and perform 3D reconstruction; obtain the 3D coordinates and spatial posture of the boom joint based on the reconstructed image and the pre-saved construction machinery joint model; Based on the 3D 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 conditions are determined; Based on the pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction factor and torque compensation are obtained according to the working condition type, operation status and slope repair angle, and the current status parameters of the construction machinery are adjusted to achieve fuel optimization.

[0007] Furthermore, based on the reconstructed image and the pre-saved engineering machinery joint model, the three-dimensional coordinates and spatial posture of the boom joint are obtained. Specifically, based on the coordinates of the key points in the image, the coordinates of the model points in the engineering machinery joint model and the intrinsic parameter matrix of the image acquisition unit, the 6-degree-of-freedom posture of the boom joint in space is obtained by solving, and the posture matrix R|t is output.

[0008] Furthermore, the pre-saved engineering machinery joint model has at least four core joints of the engineering machinery.

[0009] Furthermore, the solution is obtained as follows: using the core joint as the control point, determining the coordinates of the control point in the world coordinate system; converting the 3D point into a weighted combination of the 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 posture solution formula.

[0010] Furthermore, the current working conditions of the construction machinery are classified, and the action states and slope repair angles 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.

[0011] Furthermore, the slope repair angle is specifically: slope repair angle θ=arctan(Δz / Δx), Δz is the height difference between the bucket and the fuselage, and Δx is the horizontal projection distance.

[0012] Furthermore, the load strength factor K L Used for comprehensive reaction action state mechanical load strength, K L =(hydraulic pressure weight × current hydraulic pressure ratio) + (slope weight × slope influence factor) + (speed weight × action speed coefficient), Furthermore, the fuel supply correction coefficient is specifically: actual fuel supply = calibrated fuel supply × β × (1 + 0.02 × (current speed - economic speed)); where β is the fuel supply correction coefficient, the economic speed is the set value, and the economic speed is different under different working conditions.

[0013] Furthermore, the torque compensation is specifically: Final compensation torque ΔT = (basic compensation value × load intensity coefficient K L )+0.1×(actual torque-target torque).

[0014] A second aspect of the present invention provides a fuel efficiency optimization system based on operating condition identification, comprising: The visual recognition module is configured to: acquire an image of the environment surrounding the construction machinery; The image processing module is configured to: process the obtained environment image, generate a disparity map and perform three-dimensional reconstruction; The three-dimensional estimation module is configured to: obtain the three-dimensional coordinates and spatial posture 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 3D 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: according to the working condition type, operation status and slope repair angle, based on the pre-built fuel mapping database, obtain the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction factor and torque compensation, adjust the current state parameters of the construction machinery, and achieve fuel optimization.

[0015] Compared with the existing technology, one or more of the above technical solutions have the following beneficial effects: 1. Providing depth information through visual recognition, it addresses subtle changes that traditional sensors cannot distinguish. It also combines with existing sensors for multimodal redundancy verification to avoid false triggering. Real-time image data can be used to model the time-series motion of construction machinery based on temporal relationships, enabling more accurate understanding of the working conditions and status of construction machinery. This reduces reliance on high-precision industrial sensors (such as LiDAR), ultimately improving overall control effectiveness and reducing control costs.

[0016] 2. A pre-built fuel mapping database serves as a dynamic parameter optimization engine, digitizing 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 operating condition combination. The mapping database defines the parameter sequence for a complete action chain. For example, within the five stages of a loading operation (excavation → lifting → positioning → unloading → return), parameters transition smoothly within each stage to avoid sudden changes that increase fuel consumption.

[0017] 3. The load intensity coefficient K calculated based on the optimal parameters and the current state of the construction machinery L It's not a fixed value, but a dynamic value calculated in real time. Similarly, the fuel correction factor β not only uses a table lookup but also includes a speed compensation term, also a dynamic value. This enables global optimization of fuel efficiency at the system level. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0019] Figure 1 is a schematic diagram of the overall process of a fuel efficiency optimization method based on operating condition identification provided by one or more embodiments of the present invention; Figure 2 It is a structural diagram of a fuel efficiency optimization system based on operating condition identification provided by one or more embodiments of the present invention. DETAILED DESCRIPTION

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0022] As introduced in the background technology, existing technologies rely on manual experience to determine construction scenes and working conditions to control construction machinery. Some existing technologies use sensors combined with relevant recognition algorithms to determine construction conditions, which is prone to misjudgment and delays.

[0023] Therefore, the fuel efficiency optimization method and system based on working condition identification given in the following embodiments can accurately identify the construction working condition type (spinning, loading, leveling, slope repair) and action status (excavation, material transfer, unloading, return, etc.) in real time. Through the pre-established working condition-fuel economy mapping database, combined with the identified construction working condition type, dynamic optimization of control parameters is achieved, thereby shortening the control response delay and improving fuel efficiency.

[0024] Example 1: like Figure 1 As shown, the fuel efficiency optimization method based on operating condition identification includes the following steps: Acquire the environmental image around the construction machinery, generate a disparity map, and perform 3D reconstruction; obtain the 3D coordinates and spatial posture of the boom joint based on the reconstructed image and the pre-saved construction machinery joint model; Based on the 3D 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 conditions are determined; Based on the pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction factor and torque compensation are obtained according to the working condition type, operation status and slope repair angle, and the current status parameters of the construction machinery are adjusted to achieve fuel optimization.

[0025] The hardware configuration of this embodiment is as follows: A high-baseline binocular camera (baseline distance 12cm, f=4mm lens) is deployed on a shock-resistant bracket on the top of the cab; Embedded domain controller (computing power ≥ 10TOPS, supporting INT8 quantized inference).

[0026] The camera captures and preprocesses images of the construction machinery's surroundings. Stereo matching is used to generate a disparity map. Point cloud reconstruction and key point detection are then used to determine the 3D coordinates of the boom joints. A point-n-point pose solution is used to estimate the 3D pose. Based on the 3D pose estimate, the machine's working conditions are classified and the action state is determined. The slope correction angle is also calculated, resulting in a recognition result that is output.

[0027] The reconstructed image is used to identify the working condition using a spatiotemporal fusion algorithm. Specifically, the reconstructed image is combined with the three-dimensional coordinates of the boom joint to obtain a three-dimensional posture estimate for the construction machinery, and then the working condition label, action status, and slope repair angle are further obtained.

[0028] 3D pose estimation uses the PnP algorithm to solve the spatial pose of the boom and bucket joints (input: 2D key points + camera calibration parameters). Specifically, the PnP solver uses the coordinates of key points in the 2D image captured by the binocular camera, combined with the coordinates of the 3D model points of the construction machinery (such as an excavator) and the camera intrinsic parameter matrix, to obtain the 6-DOF pose of the boom joint in space (3D position + 3D rotation), which is the output pose matrix R|t.

[0029] This example uses HRNet keypoint detection to obtain the coordinates of 12 2D keypoints. The EPnP solver then generates the pose matrix R / t, which is then used to determine the 6-DOF pose, along with its spatial coordinates, attitude angles, and spatiotemporal features. The slope correction angle θ = arctan Δz / Δx is calculated based on the spatial coordinates and attitude angles. The ResNet-34 algorithm then performs condition classification and further status determination based on the spatiotemporal features.

[0030] The 3D pose estimation is optimized based on the existing algorithm. The specific optimization parts are shown in Table 1.

[0031] Table 1 Optimization of 3D pose estimation

[0032] The key point detection network uses a lightweight HRNet with a parameter count of <1.2M. The input image is 1920x1080. Through multi-resolution parallel processing, three types of features with resolutions of 256x144, 128x72, and 64x36 are obtained respectively. After feature fusion, a heat map is output, and key points are determined based on the heat map.

[0033] EPnP closed-form solution algorithm, specifically: (1) Definition of control points: Let the coordinates of the four control points in the world coordinate system be: ; in, 、 、 For control points j The three-dimensional coordinates in the world coordinate system.

[0034] (2) 3D point parameterization: Any 3D point P w i Expressed as a weighted combination of control points: ; α ij is the barycentric coordinate weight, i is the 3D point index (i=1, 2, ..., n).

[0035] (3) Camera projection equation: ; in, λ i is the depth scale factor, which represents the distance from the 3D point to the camera, u i 、v i are the pixel coordinates on the 2D image plane, and K is the camera intrinsic parameter matrix; K=| fx 0 u0 | K=| 0 fyv0 | K=|001|.

[0036] in, fx、fy is the focal length of the x-axis and y-axis (in pixels), u0 is the horizontal coordinate of the principal point (pixels), v0 is the vertical coordinate of the principal point (in pixels), C cj For the j The 3D coordinates of the control points in the camera coordinate system.

[0037] (4) Constructing a system of linear equations: eliminate λ i Obtain the constraint equation: ; ; The pose solution formula is as follows: ; in,[ u,v ] is the 2D image coordinate, K is the camera intrinsic parameter matrix, [R|t] is the rotation matrix + translation vector to be determined, and [X, Y, Z] is the 3D model coordinate.

[0038] The working 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] (the number of channels per group is doubled: 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 working condition classification model over ResNet-34 are shown in Table 2.

[0039] Table 2 Advantages of the working condition classification model compared to ResNet-34

[0040] The working condition classification strategy is shown in Table 3.

[0041] Table 3 Working condition classification strategy

[0042] Training strategy: The positive samples are 80,000 annotated images (20,000 for each working condition), and the negative samples are 20,000 interference scenes (rain, fog, dust, and night). The enhancement methods are dust simulation and motion blur; dust simulation is to add Gaussian noise and reduce the contrast by 30%; motion blur is a random linear blur with a kernel size of 15 pixels.

[0043] Loss function: total loss = alpha*cross entropy loss + beta*triplet loss; alpha = 0.8 to ensure basic classification accuracy; beta = 0.2 to improve feature discrimination ability.

[0044] Cross entropy loss = -Σ [true label * log(predicted probability)] Triplet loss = max(0, || anchor sample feature - positive sample feature || 2 - ||Anchor sample features- Negative sample features|| 2 + margin); where || ||2 represents the square of the Euclidean distance of the eigenvectors, and the margin is set to 1.0.

[0045] The working condition classification model is used to obtain the working condition types of construction machinery, including dumping, loading, leveling and slope repair.

[0046] (1) Working condition of throwing square: [Swing]-->Excavation and loading: bucket height <1.2m & hydraulic pressure >28MPa; Excavation and loading --> Rotation positioning: bucket height>2.0m & body rotation>30°; Rotation positioning-->Sprinkling and unloading: Bucket reaches the unloading area and opening degree>75%; Sprinkling and unloading --> Return and reset: pressure drop <15MPa & opening <20%; Return to reset --> Excavation and loading: bucket height <1.0m & pointing to the pile.

[0047] (2) Loading conditions: [Loading]-->Precision excavation: bucket height <0.8m & pressure >30MPa; Precise excavation --> Smooth lifting: bucket height > 1.5m & vibration < 0.1g; Smooth lifting --> Mining truck positioning: The outline of the mining truck is detected (visual); Truck positioning --> Accurate unloading: bucket is inside the truck bucket & height <1.0m; Accurate unloading --> efficient return: opening>85% & pressure<18MPa; Efficient return --> Precise excavation: Bucket height < 0.7m & pointing towards the pile.

[0048] A rectangular outline (length > 5m, height > 2m) & texture feature = "wheel + truck bed" was detected: it was confirmed to be a mining truck.

[0049] (3) Flat ground conditions: [Flat ground] --> Scraper grounding: bucket angle <5° & pressure <15MPa; Scraper grounding --> forward leveling: speed 0.3-0.8m / s & height fluctuation <5cm; Forward leveling --> Lifting and steering: body rotation > 20° & height > 0.3m; Lifting plate steering --> reversing positioning: the rotation angle reaches the target value ±5°; Reversing positioning --> Scraper grounding: The bucket touches the ground again; Forward leveling --> Operation completed: No elevation change for 3 consecutive times.

[0050] (4) Slope repair conditions: [Slope Repair]-->Slope Positioning: Slope detected & slope > 10°; Slope positioning --> Cutting into the slope: bucket height <0.5m & pressure >25MPa; Cut into the slope --> Slope adjustment: Move along the slope and the slope change is <1° / s; Slope trimming-->fine trimming detection: moving speed <0.2m / s; Fine-tuning inspection-->Acceptance completed: Slope error <1° for 5 seconds.

[0051] Slope repair angle calculation: θ=arctan(Δz / Δx), Δz is the height difference between the bucket and the fuselage, and Δx is the horizontal projection distance.

[0052] According to the obtained working condition type and action state, the optimized control parameters are obtained using a predetermined fuel mapping database.

[0053] The fuel control parameters corresponding to different operating conditions and action states are determined through preliminary experiments to form a fuel mapping database, as shown in Table 4.

[0054] Table 4 Fuel Mapping Database

[0055] Load intensity factor K L Used to comprehensively measure the mechanical load strength of the reaction action state.

[0056] K L =(Hydraulic pressure weight × current hydraulic pressure ratio) + (Slope weight × slope influence factor) + (Speed ​​weight × motion speed coefficient). Table 5 shows the meaning and calculation examples of each parameter during the calculation.

[0057] Table 5 Meaning of relevant parameters and calculation examples for calculating load intensity coefficient

[0058] Fuel supply correction coefficient β: actual fuel supply = calibrated fuel supply × β × (1 + 0.02 × (current speed - economic speed)).

[0059] The economic speed reference is a set value. In this embodiment, the following settings are set: swinging = 1750 rpm, loading = 1800 rpm, flat ground = 1700 rpm, and slope repair = 1900 rpm.

[0060] Torque compensation ΔT: Final compensation torque = (basic compensation value × load intensity coefficient) + 0.1 × (actual torque - target torque). The meanings of the relevant parameters are shown in Table 6.

[0061] Table 6 Explanation of the parameters of torque compensation ΔT

[0062] The above method enables precise identification of construction conditions (spinning, loading, leveling, sloping) and action states (excavating, unloading, etc.) using only cameras. Combining binocular vision 3D reconstruction with a spatiotemporal fusion algorithm, action state identification latency is less than 200ms. Visual recognition can more accurately distinguish construction conditions, offering real-time performance and more accurate recognition than traditional sensor data-based judgment. High-performance hardware ensures fast computation of the visual processing algorithm and transmits the results in real time within a 10ms cycle. Traditional algorithms rely on sensor data from the entire vehicle. For example, in the case of excavation, traditional solutions rely on handle signals and high-voltage sensor values ​​to determine whether the excavation state is in progress. Therefore, excavation can only be detected after excavation has begun, resulting in latency. However, the visual algorithm, leveraging high-performance hardware, can identify the excavation state the instant it begins and transmit the results immediately. This allows for tailored strategic responses to different working conditions, further refining the control strategy. Combined with a pre-tested fuel mapping database, engineers' discrete experience is converted into a reusable continuous parameter space, improving control effectiveness while reducing control latency.

[0063] Traditional construction machinery control is limited by human experience and subject to driver subjectivity. There is also a delay between observation, decision-making, and operation. Furthermore, the sensors in construction machinery have inherent limitations (such as misjudgment), making it difficult to identify complex movements.

[0064] This solution provides depth information through visual recognition to solve the problem of micro-slopes (such as 3° slope repair) that cannot be detected by inclination sensors.

[0065] During visual recognition, multi-modal redundancy verification is performed in combination with existing sensors. For example, during the "loading" operation, the outline of the mining truck (vision) and the sudden drop in unloading pressure (sensor) need to be detected simultaneously to avoid false triggering.

[0066] The real-time collected image data can model the sequential movements 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).

[0067] By combining the above-mentioned visual recognition with existing sensors, the working conditions and status of construction machinery can be more accurately grasped, reducing dependence 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 portable, it can quickly migrate and adapt to excavators, loaders and other equipment by updating the key point definitions of the 3D model, thereby reducing the cost of the algorithm, improving the overall control effect of construction machinery and reducing control costs. In addition, the data continuously collected during the operation of construction machinery can help to continuously optimize the algorithm.

[0068] A pre-built fuel mapping database serves as a dynamic parameter optimization engine, digitizing engineers' experience. By pre-translating discrete experience into a continuous parameter space, it eliminates the ambiguity inherent in manual operation and ensures optimal parameters for any operating condition combination. The mapping database defines the parameter sequence for a complete action chain. For example, within the five stages of a loading operation (excavation → lifting → positioning → unloading → return), parameters transition smoothly within each stage to avoid sudden changes that increase fuel consumption.

[0069] According to the optimal parameters and the current state of the construction machinery, the load intensity coefficient K is calculated. L It's not a fixed value, but a dynamic value calculated in real time. Similarly, the fuel correction factor β not only uses a table lookup but also includes a speed compensation term, also a dynamic value. This enables global optimization of fuel efficiency at the system level.

[0070] Example 2: The fuel efficiency optimization system based on operating condition identification includes: The visual recognition module is configured to: acquire an image of the environment surrounding the construction machinery; The image processing module is configured to: process the obtained environment image, generate a disparity map and perform three-dimensional reconstruction; The three-dimensional estimation module is configured to: obtain the three-dimensional coordinates and spatial posture 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 3D 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: according to the working condition type, operation status and slope repair angle, based on the pre-built fuel mapping database, obtain the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction factor and torque compensation, adjust the current state parameters of the construction machinery, and achieve fuel optimization.

[0071] Providing depth information through visual recognition, it addresses subtle changes that traditional sensors cannot distinguish. It also integrates with existing sensors for multimodal redundancy verification to prevent false triggering. Real-time image data can be used to model the time-series motion of construction machinery based on temporal relationships, enabling more accurate understanding of the machinery's operating conditions and status. This reduces reliance on high-precision industrial sensors (such as LiDAR), ultimately improving overall control effectiveness and reducing costs.

[0072] A pre-built fuel mapping database serves as a dynamic parameter optimization engine, digitizing engineers' experience. By pre-translating discrete experience into a continuous parameter space, it eliminates the ambiguity inherent in manual operation and ensures optimal parameters for any operating condition combination. The mapping database defines the parameter sequence for a complete action chain. For example, within the five stages of a loading operation (excavation → lifting → positioning → unloading → return), parameters transition smoothly within each stage to avoid sudden changes that increase fuel consumption.

[0073] The load intensity coefficient K is calculated based on the optimal parameters and the current state of the construction machinery. L It's not a fixed value, but a dynamic value calculated in real time. Similarly, the fuel correction factor β not only uses a table lookup but also includes a speed compensation term, also a dynamic value. This enables global optimization of fuel efficiency at the system level.

[0074] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall 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: The following steps are involved: Acquire the environmental image around the construction machinery, generate a disparity map, and perform 3D reconstruction; obtain the 3D coordinates and spatial posture of the boom joint based on the reconstructed image and the pre-saved construction machinery joint model; Based on the 3D 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 conditions are determined; Based on the pre-built fuel mapping database, the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction factor and torque compensation are obtained according to the working condition type, operation status and slope repair angle, and the current status parameters of the construction machinery are adjusted to achieve fuel optimization.

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

3. The fuel efficiency optimization method based on operating condition identification according to claim 2, characterized in that: The pre-saved engineering machinery joint model has at least four core joints of the engineering machinery.

4. The fuel efficiency optimization method based on operating condition identification according to claim 2, characterized in that: The solution is as follows: using the core joint as the control point, determine the coordinates of the control point in the world coordinate system; converting the 3D point 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 linear equation system to obtain the pose solution formula.

5. The fuel efficiency optimization method based on operating condition identification according to claim 1, characterized in that: Classify the current working conditions of construction machinery and determine the action status and slope repair angle under different working conditions. Specifically, obtain the classification results by combining the pre-trained working condition classification model with the set classification strategy.

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

7. The fuel efficiency optimization method based on operating condition identification according to claim 1, characterized in that: Load intensity factor K L Used for comprehensive reaction action state mechanical load strength, K L =(hydraulic pressure weight × current hydraulic pressure ratio) + (slope weight × slope influence factor) + (speed weight × action speed coefficient).

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

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

10. A system for implementing the fuel efficiency optimization method based on operating condition identification according to any one of claims 1 to 9, characterized in that: include: The visual recognition module is configured to: acquire an image of the environment surrounding the construction machinery; The image processing module is configured to: process the obtained environment image, generate a disparity map and perform three-dimensional reconstruction; The three-dimensional estimation module is configured to: obtain the three-dimensional coordinates and spatial posture 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 3D 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: according to the working condition type, operation status and slope repair angle, based on the pre-built fuel mapping database, obtain the corresponding load intensity coefficient, engine speed, main pump pressure, fuel supply correction factor and torque compensation, adjust the current state parameters of the construction machinery, and achieve fuel optimization.

Images