A method, apparatus, and storage medium for recommending work machine work method parameters

Abstract

The application provides a work method parameter recommendation method and device of a working machine and a storage medium. The method comprises: in response to monitoring that the working machine is in a target working stage, obtaining a first actual value of a target working condition parameter; based on the first actual value, matching a priori knowledge rule related to working efficiency to obtain a first recommended value of a target work method parameter; taking working efficiency as an optimization target, using a working efficiency prediction model to perform optimization within a safety boundary of the target work method parameter based on the first actual value to obtain a target candidate value of the target work method parameter; and based on the first recommended value and a second recommended value determined based on the target candidate value, determining a target recommended value of the target work method parameter. The application uses a priori knowledge rule constraint and a working efficiency prediction model optimization double-branch parallel to make the final output work method parameter recommendation value have both the working efficiency guarantee of the rule baseline and the working efficiency gain of the model optimization, thereby helping to improve the working efficiency of the working machine.

Description

Technical Field

[0001] This application relates to the field of construction machinery technology, specifically to a method, equipment, and storage medium for recommending construction machinery operation parameters. Background Technology

[0002] The complete operation process of drilling rigs typically includes multiple stages. Taking rotary drilling rigs as an example, their operation generally includes five stages: drilling, lifting, lowering, rotation, and soil dumping. At each stage, appropriate working parameters must be selected based on the working conditions of that stage to control the drilling rig to perform its operations efficiently.

[0003] Currently, the selection of construction parameters generally relies on the operator's personal experience, with the operator choosing appropriate parameters based on their perception of the current working conditions. However, there are significant differences in the accuracy of working condition perception, response speed, and the rationality of parameter selection among operators with different levels of experience, which cannot guarantee the operating efficiency of the machinery. Summary of the Invention

[0004] The main objective of this application is to propose a method, device, and storage medium for recommending operating parameters for construction machinery, aiming to improve the operating efficiency of construction machinery.

[0005] This application provides a method for recommending the working parameters of a work machine, comprising: in response to detecting that the work machine is in a target working stage, obtaining a first actual value of a target working condition parameter; the target working condition parameter is a working condition parameter of the work machine related to the target working stage; matching the first actual value with prior knowledge rules related to work efficiency to obtain a first recommended value of the target working parameter; the target working parameter is a working parameter of the work machine related to the target working stage; using work efficiency as the optimization target, using a work efficiency prediction model, optimizing within the safety boundary of the target working parameter based on the first actual value to obtain a target candidate value of the target working parameter; determining the target recommended value of the target working parameter based on the first recommended value and the second recommended value of the target working parameter; the second recommended value is determined based on the target candidate value.

[0006] In one embodiment, the step of using work efficiency as the optimization objective and employing a work efficiency prediction model to optimize within the safety boundary of the target work method parameter based on the first actual value to obtain target candidate values ​​for the target work method parameter includes: determining all candidate values ​​for the target work method parameter based on the safety boundary of the target work method parameter; for each candidate value, inputting the first actual value of the target work condition parameter and the candidate value into the work efficiency prediction model to obtain a prediction index value corresponding to the candidate value; the prediction index value includes predicted work efficiency; sorting the multiple candidate values ​​in descending order of predicted work efficiency to obtain a first sorting result for the multiple candidate values; and selecting at least one target candidate value with the highest sorting from the multiple candidate values ​​based on the first sorting result for the multiple candidate values.

[0007] In one embodiment, the predicted index value further includes a predicted fuel consumption rate; the step of selecting at least one target candidate value with a high ranking from the plurality of candidate values ​​based on the first ranking result of the plurality of candidate values ​​includes: sorting the plurality of candidate values ​​in ascending order of the predicted fuel consumption rate to obtain a second ranking result of the plurality of candidate values; for each candidate value, performing a weighted fusion of the first ranking result and the second ranking result of the candidate values ​​to obtain a target ranking result of the candidate value; and selecting at least one target candidate value with a high ranking from the plurality of candidate values ​​based on the target ranking result of the plurality of candidate values.

[0008] In one embodiment, the target working condition parameters include at least one of construction working condition parameters and operational working condition parameters; the step of obtaining a first recommended value for the target construction method parameters by matching the first actual value with prior knowledge rules related to work efficiency includes at least one of the following: matching the first actual value of the construction working condition parameters with a first prior knowledge rule to obtain the first recommended value; the first prior knowledge rule is determined based on the operational specifications of the working machinery; matching the first actual value of the operational working condition parameters with a second prior knowledge rule to obtain the first recommended value; the second prior knowledge rule is determined based on operational experience related to the working machinery.

[0009] In one embodiment, determining the target recommended value of the target construction method parameters based on the first recommended value and the second recommended value of the target construction method parameters includes: determining a third recommended value based on the product of the first recommended value and a first weighting coefficient; determining a fourth recommended value based on the product of the second recommended value and a second weighting coefficient; wherein the second weighting coefficient is determined based on the accuracy of the work efficiency prediction model in predicting work efficiency; the sum of the second weighting coefficient and the first weighting coefficient is equal to 1; and determining the target recommended value based on the sum of the third recommended value and the fourth recommended value.

[0010] In one embodiment, determining the target recommended value of the target construction method parameter based on the first recommended value and the second recommended value of the target construction method parameter includes: weighting and fusing the first recommended value and the second recommended value to obtain a fifth recommended value; and determining the target recommended value based on the comparison result between the fifth recommended value and the safety boundary of the target construction method parameter.

[0011] In one embodiment, determining the target recommended value of the target construction method parameter based on the first recommended value and the second recommended value of the target construction method parameter includes: determining the target recommended value based on the first recommended value and the second recommended value; determining the improvement in work efficiency of the target recommended value relative to the current adopted value based on the difference between the predicted work efficiency corresponding to the target recommended value and the actual work efficiency corresponding to the current adopted value of the target construction method parameter; and outputting the target recommended value and the improvement in work efficiency in response to the improvement in work efficiency being greater than 0.

[0012] In one embodiment, after determining the target recommended value of the target construction method parameters based on the first recommended value and the second recommended value of the target construction method parameters, the method further includes: in response to the operation of adopting the target recommended value, updating the work efficiency prediction model based on the first actual value, the target recommended value, and the actual work efficiency of the working machinery under the first actual value and the target recommended value; and in response to the operation of adjusting the target recommended value, updating the work efficiency prediction model based on the first actual value, the adjusted target recommended value, and the actual work efficiency of the working machinery under the first actual value and the adjusted target recommended value.

[0013] This application embodiment also provides a device for recommending the operation parameters of a work machinery, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the above-mentioned method for recommending the operation parameters of the work machinery.

[0014] This application embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the recommended method for the above-mentioned working machinery method parameters.

[0015] This application provides a method, device, and storage medium for recommending operating parameters for machinery. The method includes: in response to detecting that the machinery is in a target operating stage, obtaining a first actual value of the target operating condition parameter; matching the first actual value with prior knowledge rules related to operating efficiency to obtain a first recommended value of the target operating parameter; using operating efficiency as the optimization objective, and employing an operating efficiency prediction model, optimizing within the safety boundary of the target operating parameter based on the first actual value to obtain a target candidate value of the target operating parameter; and determining the target recommended value of the target operating parameter based on the first recommended value and a second recommended value determined based on the target candidate value. This application uses a dual-branch parallel approach of prior knowledge rule constraints and operating efficiency prediction model optimization, ensuring that the final output recommended value of the operating parameter combines the operating efficiency guarantee of the rule baseline with the operating efficiency gain of the model optimization, thereby helping to improve the operating efficiency of the machinery. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the recommended method for determining the parameters of the operating machinery provided in this application embodiment.

[0017] Figure 2 This is a schematic diagram illustrating the specific process of the recommended method for the operation machinery method parameters provided in the embodiments of this application.

[0018] Figure 3 This is a schematic diagram of the architecture of the recommended system for construction machinery and method parameters provided in the embodiments of this application.

[0019] Figure 4 This is a schematic diagram of the recommended equipment for the construction method parameters of the working machinery provided in the embodiments of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0021] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the digit " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0022] The method for recommending work machinery parameters provided in this application can be applied to a device for recommending work machinery parameters or the software of such a device. The device for recommending work machinery parameters can be the work machinery itself, a terminal, or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, or as a server cluster or distributed system composed of multiple physical servers. The software can be an application that implements the method for recommending work machinery parameters, but is not limited to the above forms.

[0023] The recommended method for the construction machinery parameters provided in this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] Please see Figure 1 The method for recommending construction machinery parameters provided in this application embodiment may include: Step S101: In response to detecting that the operating machinery is in the target operating stage, obtain the first actual value of the target operating condition parameter; the target operating condition parameter is the operating condition parameter of the operating machinery related to the target operating stage; Optionally, the operating parameters of the working machinery may include at least one of the following: construction operating parameters and operational operating parameters. Construction operating parameters can be those used to characterize the characteristics of the construction environment and the work object in which the working machinery is located, while operational operating parameters can be those used to characterize the operating status of the working machinery itself.

[0025] When the operating machinery is a rotary drilling rig, the construction condition parameters may include at least one of the following: formation type, drill bit specifications (drill bucket diameter / cylinder length / tooth type, etc.), drilling depth, and borehole diameter; the operating condition parameters may include at least one of the following: power head torque, main winch wire rope tension, and mast angle. The target operating condition parameters for different operation stages may be completely different, partially the same, or completely the same.

[0026] In practice, when the working machinery is in the target operation stage, the original values ​​of the target working condition parameters can be collected by sensors and the original values ​​can be determined as the first actual value mentioned above. Alternatively, feature extraction (such as mean feature, change amplitude feature, change frequency feature, etc.) can be performed on the original values ​​of the target working condition parameters over a period of time to obtain the feature values ​​of the target working condition parameters and the feature values ​​can be determined as the first actual value mentioned above.

[0027] Step S102: Based on the first actual value matching and prior knowledge rules related to work efficiency, obtain the first recommended value of the target construction method parameter; the target construction method parameter is the construction method parameter related to the working machinery and the target work stage; Optionally, the working method parameters of the operating machinery can be used to characterize the parameters that are actively adjusted through components such as the control panel and operating handles of the operating machinery to control the machinery to perform operating operations. When the operating machinery is a rotary drilling rig, the working method parameters may include at least one of the following: pressurization pressure, engine gear, and power head speed. The target working method parameters for different operating stages can be completely different, partially the same, or completely identical.

[0028] Optionally, the prior knowledge rules may include at least one of the following: a first prior knowledge rule determined based on the operation specifications of the operating machinery, and a second prior knowledge rule determined based on the operation experience related to the operating machinery.

[0029] Among them, the operation specifications can include the standardized operation requirements and parameter ranges specified in the operation manual and maintenance manual of the operating machinery. Most of these contents have been theoretically calculated and calibrated to verify the efficiency of the operation. Based on this, the first prior knowledge rules can be used to recommend the parameter values ​​of the construction method that guarantee the efficiency of the operation.

[0030] Operational experience can be the experience accumulated by many senior operators in long-term practice. After verification in actual operation and selection, the experience that remains is mostly the experience that has been proven to have higher operation efficiency. The second prior knowledge rules determined based on this can also be used to recommend the parameter values ​​of the construction method that guarantee operation efficiency.

[0031] Optionally, the prior knowledge rules can be represented as a set of mapping relationships between multiple sets of working condition parameter values ​​and construction method parameter values. In actual implementation, given the first actual value of the target working condition parameter, the first recommended value that has a mapping relationship with the first actual value can be matched from the prior knowledge rules using the first actual value.

[0032] Step S103: Taking work efficiency as the optimization target, using the work efficiency prediction model, optimization is performed within the safety boundary of the target construction method parameters based on the first actual value to obtain the target candidate values ​​of the target construction method parameters; Optionally, the safety boundaries of the target construction method parameters can be determined based on the safety boundary constraint rules and design thresholds related to the operating machinery and the target operation stage. When the operating machinery is a rotary drilling rig, the safety boundary constraint rules may include at least one of the following: the upper limit of the power head rotation speed is m% of the operating machinery model's design threshold, and the upper limit of the pressurization pressure is n% of the operating machinery model's design threshold; where m and n are less than 100 and can be obtained through calibration.

[0033] In practice, within the safety boundary of the target construction method parameters, a grid search algorithm or Bayesian optimization algorithm can be used to traverse all candidate values ​​of the target construction method parameters, and the operation efficiency prediction model can be called to predict the operation efficiency of each candidate value, so as to obtain at least one target candidate value with higher predicted operation efficiency.

[0034] Step S104: Based on the first recommended value and the second recommended value of the target construction method parameters, determine the target recommended value of the target construction method parameters; the second recommended value is determined based on the target candidate value.

[0035] When there is only one target candidate value, it can be directly determined as the second recommended value; when there are multiple target candidate values, the average of the multiple target candidate values ​​can be determined as the second recommended value.

[0036] After determining the second recommended value, the first and second recommended values ​​can be weighted and fused to obtain the target recommended value. Then, the target recommended value can be output on the embedded human-machine interface (e-HMI) of the machine to provide a reference for the operator to select the working method parameters.

[0037] Thus, by constraining the branches according to prior knowledge rules, we can obtain recommended values ​​for the method parameters under the current working conditions that guarantee operational efficiency within the prior knowledge rules; by optimizing the branches according to the operational efficiency prediction model, we can obtain recommended values ​​for the method parameters under the current working conditions that guarantee optimal operational efficiency within the safety boundary; the target recommended value determined based on the recommended values ​​of the two branches combines the operational efficiency guarantee of the rule baseline with the operational efficiency gain of the model optimization, which helps to improve the operational efficiency of the machinery.

[0038] Optionally, the target operating condition parameters mentioned above include at least one of the construction operating condition parameters and the operation operating condition parameters.

[0039] In one embodiment, the first recommended value of the target construction method parameter obtained in step S102 above, based on the prior knowledge rule related to the operation efficiency and the first actual value matching, includes at least one of the following: The first recommended value is obtained by matching the first actual value of the construction condition parameters with the first prior knowledge rule; the first prior knowledge rule is determined based on the operation specifications of the operating machinery. The first recommended value is obtained by matching the first actual value of the operating condition parameters with the second prior knowledge rule; the second prior knowledge rule is determined based on the operating experience related to the operating machinery.

[0040] Optionally, the first prior knowledge rule can be represented as a set of mapping relationships between multiple sets of construction condition parameter values ​​and construction method parameter values. In actual implementation, given the first actual value of the construction condition parameter, the first recommended value that has a mapping relationship with the first actual value can be matched from the first prior knowledge rule using the first actual value.

[0041] Optionally, the second prior knowledge rule can be represented as a set of mapping relationships between multiple sets of operating condition parameter values ​​and method parameter values. In actual implementation, given the first actual value of the operating condition parameter, the first recommended value that has a mapping relationship with the first actual value can be matched from the second prior knowledge rule using the first actual value.

[0042] In practice, when the first recommended value is obtained by matching the first prior knowledge rule and the second prior knowledge rule respectively, the two first recommended values ​​can be weighted and fused to obtain the final first recommended value.

[0043] Optionally, when the target method parameters include multiple method parameters, the first recommended value determined based on the first prior knowledge rule and / or the second prior knowledge rule can include the first recommended values ​​for various method parameters. For example, in clay formations using a φ1.2m drill bit, the first recommended value matched based on the first prior knowledge rule can include: engine gear recommended to be 2-3, and power head speed recommended to be 20-30 rpm. When the drilling torque fluctuation exceeds 30% and the frequency is >5 times / minute, the first recommended value matched based on the second prior knowledge rule can include: engine gear recommended to be downshifted by 1 gear to avoid stuck drill bit.

[0044] When the operating machinery is a rotary drilling rig, for its drilling, lifting, lowering, and rotation stages, the first recommended value can be obtained by matching the first actual value of the operating condition parameters and the first actual value of the construction condition parameters with a second prior knowledge rule. Correspondingly, for the drilling, lifting, lowering, and rotation stages, the second prior knowledge rule can be expressed as a set of mapping relationships between multiple sets of operating condition parameter values, construction condition parameters, and construction method parameter values.

[0045] Thus, by constraining the branches through the first and / or second prior knowledge rules, recommended values ​​for the working method parameters under the current working conditions that guarantee operational efficiency within the prior knowledge rules can be obtained, thereby helping to improve the operational efficiency of the machinery.

[0046] In one embodiment, step S103 above: taking work efficiency as the optimization objective, using a work efficiency prediction model, and optimizing within the safety boundary of the target construction method parameters based on the first actual value to obtain target candidate values ​​for the target construction method parameters, includes: Based on the safety boundary of the target construction method parameters, all candidate values ​​of the target construction method parameters are determined. For each candidate value, the first actual value of the target working condition parameter and the candidate value are input into the work efficiency prediction model to obtain the prediction index value corresponding to the candidate value; the prediction index value includes the predicted work efficiency. The candidate values ​​are sorted in descending order of predicted job efficiency to obtain the first sorting result of the candidate values. Based on the first ranking result of multiple candidate values, at least one target candidate value that ranks first is selected from the multiple candidate values.

[0047] In practice, a grid search algorithm can be used to uniformly discretize the target construction parameters within the safety boundary of the target construction parameters with a preset fixed step size to obtain all candidate values ​​of the target construction parameters; alternatively, a Bayesian optimization algorithm can be used to iteratively sample the target construction parameters multiple times within the safety boundary of the target construction parameters to obtain all candidate values ​​of the target construction parameters.

[0048] For each candidate value, the first actual value of the target working condition parameter and the candidate value of the target working method parameter can be input into the work efficiency prediction model to obtain the predicted work efficiency corresponding to the candidate value. Then, multiple candidate values ​​can be sorted in descending order of predicted work efficiency, and at least one target candidate value with the highest ranking can be selected from multiple candidate values.

[0049] Optionally, when the target construction parameters include multiple construction parameters, the candidate values ​​can be represented as a combination of candidate values; the combination of candidate values ​​can be composed of one candidate value of each of the multiple construction parameters within its respective safety boundary.

[0050] In this way, by finding the optimal branch through the operation efficiency prediction model, we can obtain the recommended values ​​of the operation method parameters with the best operation efficiency within the safety boundary under the current working conditions, which helps to improve the operation efficiency of the machinery.

[0051] In one embodiment, before inputting the first actual value and candidate value of the target operating condition parameter into the operation efficiency prediction model, the recommended method provided in this application embodiment further includes: Obtain training samples; the training samples include the second actual value of the target working condition parameters, the historical adopted value of the target working method parameters, and the actual index value of the operating machinery under the second actual value and the historical adopted value; the actual index value includes the actual operating efficiency.

[0052] The work efficiency prediction model is trained using training samples.

[0053] Optionally, the second actual value can be a historical value of the first actual value. The first actual value can be the operating condition parameter value generated by the working machine operating under the current adopted values ​​of the target working method parameters; the second actual value can be the operating condition parameter value generated by the working machine operating under the historical adopted values ​​of the target working method parameters.

[0054] Alternatively, the job efficiency prediction model can be a structured data model suitable for small to medium-sized datasets, such as gradient boosting tree (XGBoost / LightGBM), random forest, support vector regression (SVR), or deep neural network (TabNet / sTabNet).

[0055] In practice, the second actual value of the target working condition parameter, the historical adopted value of the target working method parameter, and the actual index value of the working machinery under the second actual value and the historical adopted value can be used to train the work efficiency prediction model, so that it can continuously learn the mapping relationship between the working condition parameter value, the working method parameter value and the index value, thereby improving the accuracy of the work efficiency prediction model in predicting work efficiency.

[0056] In practice, fuel consumption rate can be considered while optimizing operational efficiency, and recommended method parameter values ​​that can improve operational efficiency and reduce fuel consumption are suggested. Optionally, the above-mentioned predicted index values ​​may also include predicted fuel consumption rate; the above-mentioned actual index values ​​may also include actual fuel consumption rate.

[0057] In one embodiment, the above-mentioned selection of at least one target candidate value ranking highly from the multiple candidate values ​​based on the first ranking result of multiple candidate values ​​includes: The candidate values ​​are sorted in order of predicted fuel consumption rate from low to high to obtain a second sorting result for the candidate values. For each candidate value, the first and second sorting results of the candidate value are weighted and fused to obtain the target sorting result of the candidate value; Based on the target ranking results of multiple candidate values, at least one target candidate value that ranks first is selected from the multiple candidate values.

[0058] In practice, for each candidate value of the target working method parameter, the first actual value and the candidate value of the target working condition parameter can be input into the work efficiency prediction model to obtain the predicted work efficiency and predicted fuel consumption rate corresponding to the candidate value.

[0059] Next, the candidate values ​​can be sorted from high to low according to the predicted operating efficiency to obtain the first sorting result of the candidate values; at the same time, the candidate values ​​can be sorted from low to high according to the predicted fuel consumption rate to obtain the second sorting result of the candidate values.

[0060] Then, for each candidate value, its first sorting result and second sorting result can be weighted and fused to obtain the target sorting result of the candidate value; finally, based on the target sorting results of multiple candidate values, at least one target candidate value with the highest sorting can be selected from multiple candidate values.

[0061] In this way, while optimizing operating efficiency, fuel consumption rate is also taken into account. By optimizing the branch through the operating efficiency prediction model, recommended values ​​of operating method parameters that balance operating efficiency and fuel consumption rate within the safety boundary under the current working conditions can be obtained. This allows for improving the operating efficiency of the operating machinery while reducing its fuel consumption rate.

[0062] In one embodiment, determining the target recommended value of the target construction method parameters based on the first and second recommended values ​​of the target construction method parameters in step S104 above includes: The third recommended value is determined based on the product of the first recommended value and the first weighting coefficient. The fourth recommended value is determined based on the product of the second recommended value and the second weighting coefficient; wherein, the second weighting coefficient is determined based on the accuracy of the operation efficiency prediction model in predicting operation efficiency; the sum of the second weighting coefficient and the first weighting coefficient is equal to 1; The target recommendation value is determined based on the sum of the third and fourth recommendation values.

[0063] Optionally, after each prediction of work efficiency using the work efficiency prediction model, the predicted work efficiency can be compared with the actual work efficiency, and the accuracy of the work efficiency prediction model can be updated based on the comparison results.

[0064] In practice, we can first predict the accuracy of the work efficiency based on the work efficiency prediction model, and then determine the second weight coefficient corresponding to the accuracy. Next, we can determine the first weight coefficient by the difference between 1 and the second weight coefficient. Then, we can multiply the first recommended value by the first weight coefficient to obtain the third recommended value, and multiply the second recommended value by the second weight coefficient to obtain the fourth recommended value. Finally, we can determine the target recommended value by the third recommended value and the fourth recommended value.

[0065] Alternatively, the target recommended value can be determined using the following formula: A = B × (1 - λ) + C × λ Where A is the target recommended value, B is the first recommended value, C is the second recommended value, λ is the first weighting coefficient, and 1-λ is the second weighting coefficient.

[0066] Optionally, the first weighting coefficient λ can be in the range of 0 ≤ λ ≤ 1, where the value of λ can increase as the accuracy of the work efficiency prediction model increases. When data on the accuracy of work efficiency prediction is lacking, such as in new construction areas where the work efficiency prediction model has not yet been applied, the value of λ can be 0. In this case, a single-driven method parameter recommendation mode based on prior knowledge rule constraints can be adopted, thus achieving a cold start for method parameter recommendation. After obtaining data on the accuracy of work efficiency prediction, the method parameter recommendation mode can be switched to a dual-driven mode based on prior knowledge rule constraints and the optimization branch of the work efficiency prediction model.

[0067] Given the data on the accuracy of work efficiency prediction, the value of λ can range from 0.3 ≤ λ ≤ 0.7. When λ = 0.3, it represents a work method parameter recommendation mode dominated by prior knowledge rule constraints. When λ = 0.7, it represents a work method parameter recommendation mode dominated by the optimization branch of the work efficiency prediction model.

[0068] Thus, based on the accuracy of the work efficiency prediction model in predicting work efficiency, we can flexibly select the recommended mode of work method parameters to avoid cold start, and we can also flexibly adjust the fusion weight of the work method parameter values ​​recommended by the prior knowledge rule constraint branch and the optimization branch of the work efficiency prediction model, so that the fused target recommended value helps to improve the work efficiency of the machinery.

[0069] In one embodiment, determining the target recommended value of the target construction method parameters based on the first and second recommended values ​​of the target construction method parameters in step S104 above includes: The first and second recommendation values ​​are weighted and combined to obtain the fifth recommendation value; Based on the comparison results between the fifth recommended value and the safety boundary of the target construction method parameters, the target recommended value is determined.

[0070] In practice, the weighted fusion result of the first recommended value and the second recommended value can be determined as the fifth recommended value. Then, the fifth recommended value can be compared with the safety boundary of the target construction method parameters. If the fifth recommended value exceeds the safety boundary, the safety boundary can be determined as the target recommended value. If the fifth recommended value is within the safety boundary, the fifth recommended value can be determined as the target recommended value.

[0071] In this way, it can be ensured that the recommended values ​​of the construction method parameters do not exceed the design threshold of the machine, thereby effectively improving the operating efficiency of the machine while ensuring operational safety.

[0072] In one embodiment, determining the target recommended value of the target construction method parameters based on the first and second recommended values ​​of the target construction method parameters in step S104 above includes: Based on the first and second recommended values, the target recommended value is determined; Based on the difference between the predicted work efficiency corresponding to the target recommended value and the actual work efficiency corresponding to the current adopted value of the target method parameters, the work efficiency improvement of the target recommended value relative to the current adopted value is determined. If the improvement in work efficiency is greater than 0, output the target recommended value and the improvement in work efficiency.

[0073] In practice, after determining the target recommended values ​​for the target construction method parameters, the target recommended values ​​and the first actual values ​​of the target working condition parameters can be input into the work efficiency prediction model to obtain the predicted work efficiency corresponding to the target recommended values. Then, the predicted work efficiency corresponding to the target recommended values ​​can be compared with the actual work efficiency corresponding to the currently used values.

[0074] If the predicted work efficiency corresponding to the target recommended value is higher than the actual work efficiency corresponding to the currently adopted value, then it can be determined that the target recommended value can improve the work efficiency of the machinery relative to the currently adopted value. In this case, the target recommended value and its work efficiency improvement relative to the currently adopted value can be output on the e-HMI. The target recommended value and the currently adopted value can be displayed side-by-side on the e-HMI for easy comparison by the operator.

[0075] If the predicted work efficiency corresponding to the target recommended value is lower than or equal to the actual work efficiency corresponding to the current adopted value, it can be determined that the target recommended value cannot improve the work efficiency of the machine relative to the current adopted value. In this case, a prompt message can be output suggesting that the current adopted value be kept unchanged.

[0076] This increases the necessity of the target recommended values ​​for the output target construction method parameters and ensures that the output target recommended values ​​can effectively improve the operating efficiency of the machinery.

[0077] In one embodiment, after determining the target recommended value of the target construction method parameters based on the first and second recommended values ​​of the target construction method parameters in step S104 above, the recommendation method provided in this application embodiment further includes: In response to the operation of adopting the target recommended value, the operation efficiency prediction model is updated based on the first actual value, the target recommended value, and the actual operation efficiency of the operating machinery under the first actual value and the target recommended value; In response to the operation of adjusting the target recommended value, the work efficiency prediction model is updated based on the first actual value, the adjusted target recommended value, and the actual working efficiency of the working machinery under the first actual value and the adjusted target recommended value.

[0078] In practice, after outputting the target recommended value on the e-HMI, operations related to the target recommended value can be monitored. If an operation using the target recommended value is detected, the actual operating efficiency of the machinery under the first actual value and the target recommended value can be calculated. The training samples of the operating efficiency prediction model can be updated using the first actual value, the target recommended value, and the actual operating efficiency of the machinery under the first actual value and the target recommended value. The updated training samples can then be used to continue training the operating efficiency prediction model to update it.

[0079] If an operation to adjust the target recommended value is detected, the actual operating efficiency of the operating machinery under the first actual value and the adjusted target recommended value can be calculated. The training samples of the operating efficiency prediction model can be updated using the first actual value, the adjusted target recommended value, and the actual operating efficiency of the operating machinery under the first actual value and the adjusted target recommended value. The updated training samples can then be used to continue training the operating efficiency prediction model to update it.

[0080] If no operation related to the target recommended value is detected within the preset time period, or if an operation that ignores the target recommended value is detected, the job efficiency prediction model will remain unchanged.

[0081] In this way, the operation efficiency prediction model can be continuously iterated and optimized, so that the operation method parameter values ​​recommended by the operation efficiency prediction model can help improve the operation efficiency of the machinery.

[0082] Taking the drilling stage of a rotary drilling rig as an example, please refer to [link / reference]. Figure 2 The recommended method provided in this application embodiment may further include the following steps: Step 1 (Acquisition of Working Condition Parameters): Acquire the construction working condition parameters (such as formation type, drill string specifications, drilling depth, borehole diameter, etc.) and operational working condition parameters (such as power head torque, vibration, etc.) during the drilling stage. Among them, the formation type in the construction working condition parameters can be obtained through the operator's formation selection operation on the e-HMI (such as clay / sand / pebble / soft rock / hard rock formations, etc.), or through real-time automatic identification through on-board sensor data (such as real-time feature analysis based on power head torque-power head rotation speed-drilling speed).

[0083] Step Two (Dual-Drive Recommendation): The prior knowledge rule branch receives the drilling stage construction condition parameters and operating condition parameters, matches them with prior knowledge rules, and obtains the first recommended value for each drilling stage method parameter (such as pressurization pressure, engine gear, power head speed, etc.). The operation efficiency prediction model branch simultaneously receives the drilling stage construction condition parameters and operating condition parameters, calls the operation efficiency prediction model to predict the operation efficiency of each candidate value combination of drilling stage method parameters, and selects the top three candidate value combinations in terms of predicted operation efficiency; for each drilling stage method parameter, its value in each candidate value combination is averaged to obtain its corresponding second recommended value.

[0084] Step 3 (Weighted Fusion): The first recommended value and the second recommended value of each method parameter in the drilling stage are weighted and fused according to the weight λ, and verified by the safety boundary of each method parameter in the drilling stage (to ensure that the recommended value does not exceed the design threshold) to obtain the target recommended value of each method parameter in the drilling stage.

[0085] Step 4 (Recommendation and Interaction): Display the target recommended values ​​of each method parameter in the drilling stage and their corresponding percentage improvement in work efficiency on the e-HMI in the form of cards, so that the operator can make timely choices to "adopt" (one-click application), "ignore" (keep the current adopted value), or "adjust" (manual fine-tune) the target recommended values.

[0086] Step 5 (Feedback Loop): Record the operator selection and actual construction results (actual drilling efficiency in m / h, actual fuel consumption rate in L / m, etc.) into the construction history database. At regular intervals, the newly added historical construction data can be used to incrementally train or fully retrain the work efficiency prediction model to continuously improve its accuracy in predicting work efficiency.

[0087] The specific implementation process of this embodiment can be found in the relevant descriptions in the above embodiments, and will not be repeated here.

[0088] The method for recommending construction machinery parameters provided in this application is applicable to different stages of construction machinery operation. It employs a dual-driven fusion architecture of prior knowledge rules and a construction efficiency prediction model. By constraining the branches through prior knowledge rules, recommended values ​​with guaranteed construction efficiency within the prior knowledge rules can be obtained. By optimizing the branches through the construction efficiency prediction model, recommended values ​​with optimal construction efficiency within the safety boundary can be obtained. The two complement each other and work synergistically to effectively overcome the problem of unreliable output from pure construction efficiency prediction models under untrained geological conditions. Furthermore, a recommended output mode is adopted, presented as suggestions through an e-HMI, retaining the operator's final decision-making power and avoiding safety hazards caused by unexpected parameter mutations. In addition, the operator's adoption, ignoring, or adjustment of recommended values, as well as actual construction effect data, can be fed back to the database, forming a feedback loop to support the continuous iterative optimization of the construction efficiency prediction model.

[0089] The recommended methods provided in the embodiments of this application can be applied to, for example, Figure 3 The system recommends parameters for the operating machinery and methods shown. For example... Figure 3 As shown in the embodiments of this application, the recommendation system provided may include the following modules: an input layer, a dual-drive recommendation engine, a fusion decision layer, and a presentation layer.

[0090] The input layer can be used to input construction and operational parameters for the target work phase. Construction parameters can include the geological formation type (clay / sand / pebbles / rock, etc.) input by the operator on the e-HMI or automatically identified by the system, drill string specifications, drilling depth, and borehole diameter. Operational parameters can include parameters such as power head torque and vibration collected in real time by sensors via the CAN bus, as well as the amplitude and frequency of these parameters' changes within a preset time period.

[0091] The dual-drive recommendation engine can include a knowledge rule engine and a machine learning engine. The knowledge rule engine can embed an operation manual rule base and an experienced operator's experience base to perform deterministic condition-parameter matching inference; the machine learning engine can embed a job efficiency prediction model and a parameter optimizer to perform probabilistic job efficiency evaluation and parameter search, with the two independent channels running in parallel.

[0092] The fusion decision layer may include a weighted fusion unit, which can receive dual-channel outputs and perform weighted fusion of the recommended values ​​of the construction method parameters from the dual-channel outputs according to the λ weight. The knowledge rule engine may also embed a safety constraint library, which can be connected in series after the weighted fusion unit to independently perform safety boundary checks, ensuring that the recommended values ​​of the construction method parameters for the target operation stage do not exceed the design threshold of the operating machinery.

[0093] The presentation layer can display recommendation cards (including recommended values ​​of construction method parameters for the target operation stage, a comparison of the recommended values ​​with the currently adopted values, and the percentage increase in operation efficiency of the recommended values ​​relative to the currently adopted values) through the e-HMI interactive interface. It can also provide three operation selection buttons: "adopt / ignore / adjust". Furthermore, it can record the operator's selection in real time and feed it back to the construction history database in the machine learning engine through a RESTful interface, so as to update the operation efficiency prediction model with new construction history data in the construction history database.

[0094] It is worth mentioning that the recommendation system provided in this application embodiment can be deployed and run at the edge of the vehicle-mounted e-IPC and is available offline. In addition, the knowledge rule engine and other modules in the recommendation system can be connected through a cold start data bus. When the historical database is empty or the work efficiency prediction model is not ready, the cold start data bus can automatically bypass the machine learning engine and send the output of the knowledge rule engine directly to the weighted fusion unit (λ is forcibly set to 0), thereby realizing the ability to recommend work method parameters under zero data conditions.

[0095] This application also provides a device for recommending the parameters of a work machinery method, including a processor and a memory. The memory stores a program or instructions that can be run on the processor. When the program or instructions are executed by the processor, they implement the various steps of the above-described method embodiment for recommending the parameters of a work machinery method and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0096] Figure 4 To illustrate the hardware structure diagram of the recommended equipment for the operation machinery method parameters in this application embodiment, the recommended equipment for the operation machinery method parameters includes: The processor 401 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 402 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 402 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 402, and the processor 401 calls and executes the recommended method for the working machinery method parameters of the embodiments of this application. Input / output interface 403 is used to implement information input and output; The communication interface 404 is used to enable communication and interaction between the recommended equipment for the mechanical method parameters of this operation and other equipment. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 405 transmits information between various components of the recommended equipment for the working machinery method parameters (e.g., processor 401, memory 402, input / output interface 403, and communication interface 404). The processor 401, memory 402, input / output interface 403, and communication interface 404 are connected to each other via bus 405 to communicate within the recommended equipment for the working machinery method parameters.

[0097] The recommended equipment for the construction machinery method parameters provided in this application embodiment can realize all the steps of the above-mentioned recommended method embodiment for the construction machinery method parameters and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0098] This application also provides a storage medium storing a computer program. When the computer program is executed by a processor, it implements the various steps of the recommended method embodiment for the above-mentioned working machinery method parameters and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0099] The processor is the processor in the recommended device for the working machinery method parameters in the above embodiments. The storage medium may include a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0100] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various steps of the recommended method embodiment for the above-mentioned working machinery method parameters, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0101] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0102] This application provides a computer program product stored in a storage medium. The program product is executed by at least one processor to implement the various steps of the recommended method embodiment of the above-described working machinery method parameters, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0103] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not delete other identical elements present in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0105] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for recommending operating parameters for machinery, characterized in that, include: In response to the detection that the operating machinery is in the target operating stage, the first actual value of the target operating condition parameter is obtained; The target operating condition parameters are the operating condition parameters of the operating machinery related to the target operating stage; Based on the first actual value matching and prior knowledge rules related to work efficiency, the first recommended value of the target construction method parameter is obtained; The target construction method parameters are the construction method parameters related to the operating machinery and the target operation stage; With work efficiency as the optimization objective, the work efficiency prediction model is used to optimize within the safety boundary of the target construction method parameters based on the first actual value, thereby obtaining the target candidate values ​​of the target construction method parameters. Based on the first and second recommended values ​​of the target construction method parameters, the target recommended values ​​of the target construction method parameters are determined. The second recommended value is determined based on the target candidate value.

2. The method as described in claim 1, characterized in that, The process of optimizing the target method parameters by using an operational efficiency prediction model, based on the first actual value, within the safety boundary of the target method parameters, to obtain target candidate values ​​for the target method parameters includes: Based on the safety boundary of the target construction method parameters, all candidate values ​​of the target construction method parameters are determined. For each candidate value, the first actual value of the target working condition parameter and the candidate value are input into the work efficiency prediction model to obtain the prediction index value corresponding to the candidate value; the prediction index value includes the predicted work efficiency. The candidate values ​​are sorted in descending order of predicted job efficiency to obtain a first sorting result for the candidate values. Based on the first ranking result of the plurality of candidate values, at least one target candidate value that ranks first is selected from the plurality of candidate values.

3. The method as described in claim 2, characterized in that, The predicted index value also includes the predicted fuel consumption rate; the step of selecting at least one target candidate value with the highest ranking from the plurality of candidate values ​​based on the first ranking result of the plurality of candidate values ​​includes: The candidate values ​​are sorted in ascending order of predicted fuel consumption rate to obtain a second sorting result for the candidate values. For each candidate value, the first sorting result and the second sorting result of the candidate value are weighted and fused to obtain the target sorting result of the candidate value; Based on the target ranking results of the multiple candidate values, at least one target candidate value with the highest ranking is selected from the multiple candidate values.

4. The method as described in claim 1, characterized in that, The target operating condition parameters include at least one of the construction operating condition parameters and the operational operating condition parameters; The first recommended value for the target construction method parameter, obtained by matching the first actual value with prior knowledge rules related to work efficiency, includes at least one of the following: The first recommended value is obtained by matching the first actual value of the construction condition parameters with the first prior knowledge rule. The first prior knowledge rule is determined based on the operating specifications of the machinery. The first recommended value is obtained by matching the first actual value of the operating condition parameters with the second prior knowledge rule. The second prior knowledge rule is determined based on operational experience related to the machinery being operated.

5. The method as described in claim 1, characterized in that, The step of determining the target recommended value of the target construction method parameters based on the first recommended value and the second recommended value of the target construction method parameters includes: The third recommended value is determined based on the product of the first recommended value and the first weighting coefficient. A fourth recommended value is determined based on the product of the second recommended value and the second weighting coefficient; wherein the second weighting coefficient is determined based on the accuracy of the work efficiency prediction model in predicting work efficiency; the sum of the second weighting coefficient and the first weighting coefficient is equal to 1; The target recommendation value is determined based on the sum of the third recommendation value and the fourth recommendation value.

6. The method as described in claim 1, characterized in that, The step of determining the target recommended value of the target construction method parameters based on the first recommended value and the second recommended value of the target construction method parameters includes: The first recommendation value and the second recommendation value are weighted and fused to obtain the fifth recommendation value; The target recommended value is determined based on the comparison between the fifth recommended value and the safety boundary of the target construction method parameters.

7. The method as described in claim 1, characterized in that, The step of determining the target recommended value of the target construction method parameters based on the first recommended value and the second recommended value of the target construction method parameters includes: The target recommendation value is determined based on the first recommendation value and the second recommendation value; Based on the difference between the predicted work efficiency corresponding to the target recommended value and the actual work efficiency corresponding to the current adopted value of the target method parameter, the work efficiency improvement of the target recommended value relative to the current adopted value is determined. In response to the fact that the improvement in work efficiency is greater than 0, the target recommended value and the improvement in work efficiency are output.

8. The method as described in claim 1, characterized in that, After determining the target recommended value of the target construction method parameters based on the first and second recommended values ​​of the target construction method parameters, the method further includes: In response to the operation of adopting the target recommended value, the operation efficiency prediction model is updated based on the first actual value, the target recommended value, and the actual operation efficiency of the operating machinery under the first actual value and the target recommended value; In response to the operation of adjusting the target recommended value, the operation efficiency prediction model is updated based on the first actual value, the adjusted target recommended value, and the actual operation efficiency of the operating machinery under the first actual value and the adjusted target recommended value.

9. A device for recommending operating parameters for machinery, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements the method as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 8.

Images