Excavator crushing condition recognition model training method, recognition method and device

Abstract

The invention provides an excavator crushing condition recognition model training method, recognition method and device, and can be applied to the technical field of excavators. The training method comprises the following steps: acquiring training data; wherein the training data comprises first sample features correspondingly extracted based on the first historical working condition data, second sample features correspondingly extracted based on the second historical working condition data and third sample features correspondingly extracted based on the third historical working condition data; training the machine learning model based on the training data and the working condition label corresponding to each sample feature in the training data to obtain a crushing working condition identification model; wherein the working condition label is used for marking whether the sample excavator is in a crushing working condition. According to the method, whether the excavator is in the crushing working condition or not can be accurately recognized by training the crushing working condition recognition model, and the accuracy of excavator crushing working condition recognition is effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of excavators, and in particular to a training method, an identification method and a device for identifying a crushing condition of an excavator. Background Art

[0002] An excavator is a kind of engineering machinery used for earthwork construction. It is widely used in mining, construction, road, water conservancy and other engineering fields. It has a wide range of applications and high work efficiency. Among them, the crushing condition of the excavator is one of its important application scenarios, which is mainly used in the fields of building demolition, mining and road construction.

[0003] Because the excavator is in the crushing condition, the wear of the key parts of the excavator working device will be greater than that in the general working condition. At the same time, during the crushing operation, the oil temperature of the excavator hydraulic system rises quickly. When the oil temperature is too high, the seals are also prone to aging and damage. Therefore, in order to reasonably arrange the maintenance plan of the excavator equipment and extend the service life of the excavator equipment, it is necessary to identify whether the excavator is in the crushing condition.

[0004] In the prior art, it is generally determined whether the excavator is in a crushing condition by visual recognition technology or by rules such as whether the pressure data of the excavator meets the threshold conditions. However, visual recognition technology is easily affected by the environment. For example, weather and dust will greatly affect the accuracy of recognition. There may be errors if the pressure value is completely relied upon to judge the crushing condition, because the pressure value may be affected by many factors, such as leakage of the hydraulic system, performance changes of the pump, etc. It is driven by rules such as thresholds and lacks the ability to generalize complex scenarios. Different rule thresholds need to be set for different models, and unified thresholds are prone to misjudgment. Therefore, the existing excavator crushing condition method has the problem of low recognition accuracy. Summary of the invention

[0005] The present application provides a model training method, an identification method and a device for identifying crushing conditions of an excavator, so as to solve the technical problem of low accuracy in identifying crushing conditions of an excavator in the prior art.

[0006] According to the first aspect disclosed in the present application, the present application provides a method for training an excavator crushing condition identification model, comprising:

[0007] Acquire training data; wherein the training data includes first sample features correspondingly extracted based on each first historical operating condition data, second sample features correspondingly extracted based on each second historical operating condition data, and third sample features correspondingly extracted based on each third historical operating condition data; the first historical operating condition data includes a plurality of pilot pressures collected from a first sample excavator with a crushing-only mode within a historical accumulated working time interval, the second historical operating condition data includes a plurality of dual-pump pressure sums collected from a second sample excavator of a dual-pump confluence type within a historical accumulated working time interval, and the third historical operating condition data includes a plurality of dual-pump pressure differences collected from a third sample excavator of a non-dual-pump confluence type within a historical accumulated working time interval;

[0008] Based on the training data and the working condition labels corresponding to each sample feature in the training data, the machine learning model is trained to obtain a crushing working condition recognition model; wherein the working condition label is used to mark whether the sample excavator is in a crushing working condition.

[0009] In a feasible implementation manner, the first sample characteristics include a pressure mean value, a pressure change rate, and a pressure change amplitude of the pilot pressure.

[0010] In a feasible implementation manner, the second sample feature includes a pressure sum mean, a pressure sum standard deviation, and a pressure sum change rate of the dual-pump pressure sum.

[0011] In a feasible implementation manner, the third sample feature includes a pressure difference mean, a pressure difference standard deviation, and a pressure difference change rate of the dual-pump pressure difference.

[0012] In a feasible implementation manner, the sample features in the training data further include frequency domain features, and the frequency domain features are obtained by the following method:

[0013] The historical operating condition data is converted into frequency domain data, and frequency domain features are extracted based on the frequency domain data.

[0014] According to a second aspect disclosed in the present application, the present application provides a method for identifying a crushing condition of an excavator, comprising:

[0015] Acquire current working condition data of the target excavator to be identified; wherein the current working condition data includes a plurality of pilot pressures, a plurality of dual-pump pressures, and a plurality of dual-pump pressure differences collected within a current accumulated working time interval;

[0016] Extracting a first identification feature based on the pilot pressure in the current operating condition data, extracting a second identification feature based on the dual pump pressure in the current operating condition data, and extracting a third identification feature based on the dual pump pressure difference in the current operating condition data;

[0017] Input the first identification feature, the second identification feature and the third identification feature into a crushing condition identification model respectively, and obtain corresponding identification results respectively; wherein the crushing condition identification model is obtained by training according to the excavator crushing condition identification model training method described in the first aspect;

[0018] If there is at least one identification result indicating that the target excavator is in a crushing working condition, it is determined that the target excavator is in a crushing working condition.

[0019] According to the third aspect disclosed in the present application, the present application provides an excavator crushing condition identification model training device, comprising:

[0020] A training data acquisition module, used to acquire training data; wherein the training data includes first sample features correspondingly extracted based on each first historical operating condition data, second sample features correspondingly extracted based on each second historical operating condition data, and third sample features correspondingly extracted based on each third historical operating condition data; the first historical operating condition data includes a plurality of pilot pressures collected by a first sample excavator with a crushing-only mode within a historical accumulated working time interval, the second historical operating condition data includes a plurality of dual-pump pressure sums collected by a second sample excavator of a dual-pump confluence type within a historical accumulated working time interval, and the third historical operating condition data includes a plurality of dual-pump pressure differences collected by a third sample excavator of a non-dual-pump confluence type within a historical accumulated working time interval;

[0021] The recognition model training module is used to train the machine learning model based on the training data and the working condition labels corresponding to each sample feature in the training data to obtain a crushing working condition recognition model; wherein the working condition label is used to mark whether the sample excavator is in a crushing working condition.

[0022] According to a fourth aspect disclosed in the present application, the present application provides an excavator crushing condition identification device, comprising:

[0023] An identification data acquisition module is used to obtain the current working condition data of the target excavator to be identified; wherein the current working condition data includes a plurality of pilot pressures and a plurality of dual pump pressures and a plurality of dual pump pressure differences collected within a current accumulated working time interval;

[0024] an identification feature extraction module, configured to extract a first identification feature based on the pilot pressure in the current operating condition data, extract a second identification feature based on the dual pump pressure in the current operating condition data, and extract a third identification feature based on the dual pump pressure difference in the current operating condition data;

[0025] A crushing condition identification module is used to input the first identification feature, the second identification feature and the third identification feature into a crushing condition identification model respectively, and obtain corresponding identification results respectively; wherein the crushing condition identification model is obtained by training according to the excavator crushing condition identification model training method described in the first aspect; if there is at least one identification result characterizing that the target excavator is in a crushing condition, it is determined that the target excavator is in a crushing condition.

[0026] According to a fifth aspect disclosed in the present application, the present application provides an electronic device, comprising a processor, and a memory communicatively connected to the processor;

[0027] The memory stores computer-executable instructions;

[0028] The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of the first aspect or the second aspect.

[0029] According to a sixth aspect disclosed in the present application, the present application provides an excavator, which includes the electronic device described in the fifth aspect.

[0030] According to the seventh aspect disclosed in the present application, the present application provides a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed, they are used to implement the method described in any one of the first aspect or the second aspect.

[0031] According to an eighth aspect disclosed in the present application, the present application provides a computer program product, including a computer program, and when the computer program is executed, it is used to implement the method in any one of the first aspect or the second aspect.

[0032] Compared with the prior art, this application has the following beneficial effects:

[0033] The present application provides a training method, identification method and device for an excavator crushing working condition identification model. By extracting the feature data of the original working condition data of the excavator, a comprehensive and targeted combination feature is constructed to ensure the significant distinction between crushing working conditions and non-crushing working conditions. The constructed sample features are combined with sample labels to train a machine learning model to obtain a crushing working condition identification model. An automated and intelligent working condition identification solution is provided to achieve accurate identification of the excavator crushing working condition and improve the accuracy of the excavator crushing working condition identification.

[0034] At the same time, differentiated feature combinations are constructed for different excavator types, solving the adaptation problems caused by the diversity of excavator types and different installation methods of breaker hammers. Flexible feature construction solutions are provided for different excavator types and installation methods to adapt to various working conditions. It has strong adaptability to unknown equipment and wide applicability, and realizes accurate identification of the crushing working conditions of excavators of unknown equipment types. It is suitable for various excavator types and complex working conditions, and has good versatility and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0036] Figure 1 A flow chart of a method for training a model for identifying crushing conditions of an excavator provided in an embodiment of the present application;

[0037] Figure 2 A schematic flow chart of a method for identifying a crushing condition of an excavator provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of the structure of an excavator crushing condition identification model training device provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of the structure of a device for identifying crushing conditions of an excavator provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0041] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0042] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0043] An excavator is a kind of engineering machinery used for earthwork construction. It is widely used in mining, construction, road, water conservancy and other engineering fields. It has a wide range of applications and high work efficiency. Among them, the crushing condition of the excavator is one of its important application scenarios, which is mainly used in the fields of building demolition, mining and road construction.

[0044] Because the impact force of the excavator's working device (such as the boom, dipper and breaker) is much greater than that of ordinary excavation conditions, the wear of the key parts of the excavator's working device will be greater in the crushing condition than in the general working condition. At the same time, during the crushing operation, the oil temperature of the excavator's hydraulic system rises quickly (because the frequent impact of the breaker will cause the hydraulic oil to generate a lot of heat in a short time). When the oil temperature is too high, the seals are also prone to aging and damage, resulting in leakage of the hydraulic system or accelerated deterioration of the hydraulic oil.

[0045] Therefore, it is necessary to conduct statistical analysis on the working state of the excavator in the crushing condition in order to reasonably arrange the maintenance plan of the excavator equipment and extend the service life of the excavator equipment. The premise of conducting statistical analysis on the working state of the excavator in the crushing condition is to identify whether the excavator is in the crushing condition.

[0046] In the prior art, it is generally determined whether the excavator is in a crushing condition by visual recognition technology or by rules such as whether the pressure data of the excavator meets the threshold conditions. However, visual recognition technology is easily affected by the environment, such as weather and dust, which will greatly affect the accuracy of recognition. There may be errors in relying entirely on pressure values ​​to judge the crushing condition, because the pressure value may be affected by many factors, such as leakage of the hydraulic system, performance changes of the pump, etc. It is based on rule-driven methods such as thresholds and lacks the ability to generalize complex scenarios. Different rule thresholds need to be set for different models, and unified thresholds are prone to misjudgment.

[0047] Therefore, the existing methods for identifying the crushing working conditions of excavators have the problem of low recognition accuracy. In addition, excavators of different brands and models may have differences in hydraulic system design and performance, and the existing recognition methods may not be fully compatible with all models of excavators.

[0048] The following is a detailed description of the technical solutions of the excavator crushing condition identification model training method and the excavator crushing condition identification method provided by the present application through specific embodiments. It should be noted that the following embodiments can exist independently or in combination with each other, and the same or similar contents may not be repeated in different embodiments.

[0049] Figure 1 A flow chart of a method for training a model for identifying crushing conditions of an excavator provided in an embodiment of the present application is shown in FIG. Figure 1In some embodiments, the process of the excavator crushing condition identification model training method includes the following steps:

[0050] S101, obtaining training data; wherein the training data includes first sample features extracted based on each first historical operating condition data, second sample features extracted based on each second historical operating condition data, and third sample features extracted based on each third historical operating condition data; the first historical operating condition data includes a plurality of pilot pressures collected from a first sample excavator with a crushing-only mode within a historical accumulated working time interval, the second historical operating condition data includes a plurality of dual-pump pressure sums collected from a second sample excavator of a dual-pump confluence type within a historical accumulated working time interval, and the third historical operating condition data includes a plurality of dual-pump pressure differences collected from a third sample excavator of a non-dual-pump confluence type within a historical accumulated working time interval.

[0051] Among them, according to the different types of excavators, the excavators are divided into the first excavator with a special crushing mode, the second excavator of the dual-pump confluence type, and the third excavator without dual-pump confluence. According to the different types of excavators, different working condition data of the excavators are collected, and different data features are extracted to generate differentiated feature combinations, so as to construct multi-feature combinations driven by multi-dimensional data for different types of excavators as training data, so that the crushing condition recognition model completed by subsequent training can cover the crushing condition recognition of different types of excavators, and improve the compatibility of the crushing condition recognition model. At the same time, the training data constructed by multi-dimensional feature extraction can effectively enhance the robustness of the crushing condition recognition model and improve its accuracy of crushing condition recognition.

[0052] Specifically, the pilot pressure is the pressure acting on the oil port of the hydraulic pilot valve, which is used to reflect the action strength of the pilot valve. Generally, when the pilot pressure is continuously greater than a certain threshold, the excavator is in a crushing condition.

[0053] Specifically, the double pump pressure sum is the sum of the left pump pressure and the right pump pressure in the excavator main pump. Generally, when the double pump pressure sum is continuously greater than a certain threshold, the excavator is in a crushing condition. The double pump pressure sum satisfies the following formula:

[0054]

[0055] in, Indicates the dual pump pressure and, Indicates the left pump pressure, Indicates right pump pressure.

[0056] Specifically, the dual-pump pressure difference is the difference between the left pump pressure and the right pump pressure in the excavator main pump. When it is continuously greater than a certain threshold, the excavator is in the crushing condition. Among them, the dual-pump pressure difference satisfies the following formula

[0057]

[0058] in, Indicates the pressure difference between the two pumps. Indicates the left pump pressure, Indicates right pump pressure.

[0059] Specifically, the accumulated working hours are the working hours that the excavator is in working condition, and the accumulated working hours interval is the time interval from when the excavator starts working to when it finishes working.

[0060] Specifically, dual-pump confluence is a flow control technology that improves the working efficiency of the hydraulic system by confluence of the main pump and the auxiliary pump. Dual-pump confluence type excavators use dual-pump confluence technology to combine the output flow of the two pumps under specific working conditions to meet greater flow requirements. Non-dual-pump confluence type excavators usually only use one main pump or two independent pumps without confluence. Each pump works independently to provide the required flow and pressure for different parts of the excavator.

[0061] Preferably, the first sample characteristics include a pressure mean value, a pressure change rate and a pressure change amplitude of the pilot pressure.

[0062] Among them, the pilot pressure is Indicates that the mean pressure of the pilot pressure is Used to reflect the action strength of the pilot valve and the standard deviation of the pilot pressure Used to characterize the change amplitude of the pilot pressure within the interval and the pressure change rate of the pilot pressure Used to capture the frequency of joystick movements. Specifically, Indicates the pressure difference of the pilot pressure at adjacent time points, Represents the corresponding time interval. Organized in vector form, the first sample features are obtained.

[0063] Preferably, the second sample feature includes a pressure sum mean, a pressure sum standard deviation and a pressure sum change rate of the dual pump pressure sum.

[0064] Among them, the dual pump pressure and Indicates that It represents the mean value of the pressure and pressure of the dual pumps. The pressure and standard deviation of the dual pump pressure are represented by, Indicates the pressure and change rate of the dual pump pressure. Specifically, It represents the pressure difference between the double pump pressures at adjacent time points. Represents the corresponding time interval. Organized in vector form, the second sample features are obtained.

[0065] Preferably, the third sample feature includes a pressure difference mean value, a pressure difference standard deviation and a pressure difference change rate of the dual pump pressure difference.

[0066] Among them, the pressure difference between the two pumps is Indicates that Indicates the mean pressure difference of the dual pump pressure difference, The standard deviation of the pressure difference representing the pressure difference between the two pumps, Indicates the pressure difference change rate of the dual pump pressure difference. Specifically, The pressure difference value representing the pressure difference between the two pumps at adjacent time points, Represents the corresponding time interval. Organized in vector form, the third sample feature is obtained.

[0067] Preferably, the sample features in the training data also include frequency domain features, and the frequency domain features are obtained by the following method: converting the historical operating condition data into frequency domain data, and extracting the frequency domain features based on the frequency domain data.

[0068] Among them, wavelet transform or Fourier transform is used to analyze the spectral characteristics of the working condition data, extract the energy distribution of high-frequency vibration in the crushing condition, and combine the time domain characteristics (such as pressure mean and change rate) and frequency domain characteristics (such as spectrum peak) to construct a feature vector. By adding frequency domain features, the high-frequency dynamic characteristics implicit in the crushing condition can be captured, and the robustness to high-noise data is stronger.

[0069] Specifically, multiple sensors (main pump pressure sensor, pilot pressure sensor, and cumulative working time timer, etc.) are installed on the excavator to collect relevant data of the excavator, and the collected data is sent to the local controller (ECU) or remote server through the CAN bus or wireless module.

[0070] Preferably, before extracting features from the historical operating condition data, the data may be preprocessed, such as performing denoising and normalization processing on the collected data.

[0071] Preferably, the operating condition data can be screened by collecting the excavator engine speed, so as to exclude abnormal operating condition data according to the engine load state obtained according to the engine speed.

[0072] S102, based on the training data and the working condition labels corresponding to each sample feature in the training data, the machine learning model is trained to obtain a crushing working condition recognition model; wherein the working condition label is used to mark whether the sample excavator is in a crushing working condition.

[0073] Among them, the machine learning model is trained using sample features with working condition labels, and the model parameters are continuously adjusted to iteratively optimize the objective function until a recognition model is obtained that can accurately identify the crushing conditions of different types of excavators.

[0074] Specifically, the objective function satisfies the following formula:

[0075]

[0076] Among them, L represents the loss function, f represents the machine learning model, and F represents the parameters of the machine learning model. represents the parameters of the loss function, and Y represents the working condition label.

[0077] Preferably, before training the machine learning model, feature selection can be performed on the training data, and the feature combination that contributes most to the classification of the crushing conditions can be selected using a feature importance analysis method (such as information gain or SHAP value).

[0078] Preferably, after the crushing condition recognition model is trained, the model can also be verified and optimized, for example, by using a verification set to evaluate the model performance (indicators such as accuracy and recall), and by improving the model performance through hyperparameter optimization (such as grid search).

[0079] Specifically, the working condition labels used to annotate sample features can be automatically generated based on the annotations in the equipment logs and combined with the excavator operation logs (such as the breaker start-up records), or they can be manually annotated by experts and technicians. Represents the i-th working condition label, 0 represents non-crushing working condition, and 1 represents crushing working condition.

[0080] Specifically, the machine learning model includes support vector machine, random forest or deep learning model, etc. Among them, the deep learning model includes convolutional neural network (CNN), long short-term memory network (LSTM) or hybrid model (such as CNN-LSTM), etc.

[0081] Specifically, after the training of the crushing condition recognition model is completed, the trained crushing condition recognition model is deployed on the ECU, edge computing devices, and cloud database for real-time identification of the target excavator condition.

[0082] In this embodiment, by extracting the feature data of the original working condition data of the excavator, a comprehensive and targeted combination feature is constructed to ensure the significant distinction between the crushing condition and the non-crushing condition, and the constructed sample features are combined with sample labels to train the machine learning model to obtain the crushing condition recognition model. An automated and intelligent working condition recognition solution is provided to achieve accurate recognition of the excavator's crushing condition and improve the accuracy of the excavator's crushing condition recognition.

[0083] At the same time, differentiated feature combinations are constructed for different excavator types, solving the adaptation problems caused by the diversity of excavator types and different installation methods of breaker hammers. Flexible feature construction solutions are provided for different excavator types and installation methods to adapt to various working conditions. It has strong adaptability to unknown equipment and wide applicability, and realizes accurate identification of the crushing working conditions of excavators of unknown equipment types. It is suitable for various excavator types and complex working conditions, and has good versatility and scalability.

[0084] Figure 2 A schematic diagram of a method for identifying a crushing condition of an excavator provided in an embodiment of the present application, see Figure 2 In some embodiments, the process of the excavator crushing condition identification method includes the following steps:

[0085] S201, obtaining current working condition data of the target excavator to be identified; wherein the current working condition data includes a plurality of pilot pressures, a plurality of dual-pump pressure sums, and a plurality of dual-pump pressure differences collected within a current accumulated working time interval.

[0086] The current working condition data of the target excavator to be identified is collected for subsequent extraction of feature data for identification of the crushing working condition identification model based on the working condition data.

[0087] Specifically, current working condition data is collected through sensors installed on the target excavator.

[0088] S202, extracting a first identification feature based on the pilot pressure in the current operating condition data, extracting a second identification feature based on the dual pump pressures in the current operating condition data, and extracting a third identification feature based on the dual pump pressure difference in the current operating condition data.

[0089] Among them, in the case of an unknown target excavator type, different recognition feature combinations are constructed based on the current working condition data to ensure that the crushing condition recognition model can identify whether the target excavator is in a crushing condition.

[0090] S203, inputting the first recognition feature, the second recognition feature and the third recognition feature into the crushing condition recognition model respectively, and obtaining corresponding recognition results respectively; wherein the crushing condition recognition model is obtained by training according to the excavator crushing condition recognition model training method.

[0091] After the three sets of identification features are input into the crushing condition identification model, the model will output three sets of corresponding identification results. .

[0092] S204: If there is at least one identification result indicating that the target excavator is in a crushing condition, it is determined that the target excavator is in a crushing condition.

[0093] Among the three sets of recognition results, if there is at least one set of recognition results , it indicates that the target excavator is in crushing condition.

[0094] Preferably, if there is no identification result indicating that the target excavator is in a crushing condition, it is determined that the target excavator is not in a crushing condition.

[0095] Among the three sets of recognition results, if the three sets of recognition results , it indicates that the target excavator is not in the crushing condition.

[0096] In this embodiment, for an excavator of unknown equipment type, the crushing condition and the non-crushing condition can be automatically identified by combining the working condition data features, which effectively improves the accuracy of the excavator's crushing condition identification.

[0097] Preferably, all collected data, feature data and identification results of the target excavator are also recorded through a cloud platform or a local storage device for statistical analysis of the data, visualization of historical data of different working conditions, and fault tracing, and analysis of crushing equipment abnormalities in combination with the working condition identification results.

[0098] Figure 3 is a structural schematic diagram of an excavator crushing condition identification model training device provided in an embodiment of the present application, see Figure 3 The excavator crushing working condition identification model training device includes various functional modules for implementing the aforementioned excavator crushing working condition identification model training method, and any functional module can be implemented by software and / or hardware.

[0099] In some embodiments, the excavator crushing condition recognition model training device 300 includes a training data acquisition module 301 and a recognition model training module 302. Among them:

[0100] The training data acquisition module 301 is used to acquire training data; wherein the training data includes first sample features correspondingly extracted based on each first historical working condition data, second sample features correspondingly extracted based on each second historical working condition data, and third sample features correspondingly extracted based on each third historical working condition data; the first historical working condition data includes a plurality of pilot pressures collected by a first sample excavator with a crushing-specific mode within a historical accumulated working time interval, the second historical working condition data includes a plurality of dual-pump pressure sums collected by a second sample excavator of a dual-pump confluence type within a historical accumulated working time interval, and the third historical working condition data includes a plurality of dual-pump pressure differences collected by a third sample excavator of a non-dual-pump confluence type within a historical accumulated working time interval;

[0101] The recognition model training module 302 is used to train the machine learning model based on the training data and the working condition labels corresponding to each sample feature in the training data to obtain a crushing working condition recognition model; wherein the working condition label is used to mark whether the sample excavator is in a crushing working condition.

[0102] In some embodiments, the first sample characteristic includes a pressure mean value, a pressure change rate, and a pressure change amplitude of the pilot pressure.

[0103] In some embodiments, the second sample characteristics include a pressure sum mean, a pressure sum standard deviation, and a pressure sum change rate of the dual pump pressure sum.

[0104] In some embodiments, the third sample feature includes a pressure difference mean, a pressure difference standard deviation, and a pressure difference change rate of the dual-pump pressure difference.

[0105] In some embodiments, the sample features in the training data also include frequency domain features, and the training data acquisition module 301 is further used to:

[0106] The historical operating condition data is converted into frequency domain data, and frequency domain features are extracted based on the frequency domain data.

[0107] The excavator crushing condition identification model training device 300 provided in the embodiment of the present application is used to execute the technical solution provided in the aforementioned excavator crushing condition identification model training method embodiment. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment and will not be repeated here.

[0108] Figure 4 is a schematic diagram of a structure of an excavator crushing condition identification device provided in an embodiment of the present application, see Figure 4 The excavator crushing working condition identification device includes various functional modules for implementing the aforementioned excavator crushing working condition identification method, and any functional module can be implemented by software and / or hardware.

[0109] In some embodiments, the excavator crushing condition identification device 400 includes an identification data acquisition module 401, an identification feature extraction module 402 and a crushing condition identification module 403. Among them:

[0110] The identification data acquisition module 401 is used to obtain the current working condition data of the target excavator to be identified; wherein the current working condition data includes a plurality of pilot pressures and a plurality of dual pump pressures and a plurality of dual pump pressure differences collected within the current accumulated working time interval;

[0111] The identification feature extraction module 402 is used to extract a first identification feature based on the pilot pressure in the current working condition data, extract a second identification feature based on the dual pump pressure in the current working condition data, and extract a third identification feature based on the dual pump pressure difference in the current working condition data;

[0112] The crushing condition identification module 403 is used to input the first identification feature, the second identification feature and the third identification feature into the crushing condition identification model respectively, and obtain corresponding identification results respectively; wherein the crushing condition identification model is obtained by training according to the excavator crushing condition identification model training method; if there is at least one identification result characterizing that the target excavator is in a crushing condition, it is determined that the target excavator is in a crushing condition.

[0113] In some embodiments, the crushing condition identification module 403 is further used to:

[0114] If there is no identification result indicating that the target excavator is in a crushing condition, it is determined that the target excavator is not in a crushing condition.

[0115] The excavator crushing condition identification device 400 provided in the embodiment of the present application is used to execute the technical solution provided in the aforementioned excavator crushing condition identification method embodiment. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment and will not be repeated here.

[0116] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements, or all in the form of hardware, or some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the crushing condition identification module 403 can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The function of the above crushing condition identification module 403. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in software form.

[0117] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, see Figure 5 The electronic device 500 includes a processor 501 and a memory 02 communicatively connected to the processor 501;

[0118] Memory 502 stores computer executable instructions;

[0119] The processor 501 executes the computer execution instructions stored in the memory 502 to implement the technical solution of the aforementioned excavator crushing working method.

[0120] In the above-mentioned electronic device 500, the memory 502 and the processor 501 are electrically connected directly or indirectly to realize data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines, such as through a bus. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc., but it does not mean that there is only one bus or one type of bus. The memory 502 stores computer execution instructions for implementing the above-mentioned excavator crushing working condition method, including at least one software function module that can be stored in the memory 502 in the form of software or firmware. The processor 501 executes various functional applications and data processing by running the software programs and modules stored in the memory 502.

[0121] The memory 502 includes at least one type of readable storage medium, including but not limited to random access memory (RAM), read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), and electrically erasable read-only memory (EEPROM). The memory 502 is used to store programs, and the processor 501 executes the programs after receiving the execution instruction. Furthermore, the software programs and modules in the above-mentioned memory 502 may also include an operating system, which may include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide an operating environment for other software components.

[0122] The processor 501 may be an integrated circuit chip having the ability to process signals. The above-mentioned processor 501 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), etc. The disclosed methods, steps and logic block diagrams in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor, or the processor 501 may be any conventional processor, etc.

[0123] The electronic device 500 is used to execute the technical solution provided by the aforementioned excavator crushing condition identification model training method or the excavator crushing condition identification method embodiment. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment and will not be repeated here.

[0124] An embodiment of the present application also provides an excavator, which includes the aforementioned electronic device.

[0125] The above-mentioned excavator trains the crushing condition recognition model deployed in the electronic equipment, extracts the characteristic data of the excavator's original condition data, and accurately recognizes the excavator's crushing condition, thereby improving the accuracy of the excavator's crushing condition recognition.

[0126] Specifically, a plurality of sensors are installed on the excavator, such as a main pump pressure sensor, a pilot pressure sensor and a cumulative working timer, etc., to collect relevant working condition data of the excavator.

[0127] An embodiment of the present application also provides a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed, they are used to implement the technical solution of the aforementioned excavator crushing condition identification model training method or excavator crushing condition identification method.

[0128] The computer-readable storage medium mentioned above can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The computer-readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0129] An exemplary readable storage medium is coupled to the processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the excavator crushing condition identification model training device or the excavator crushing condition identification device control device.

[0130] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, is used to implement the technical solution of the aforementioned excavator crushing condition identification model training method or excavator crushing condition identification method.

[0131] In the above embodiments, those skilled in the art will appreciate that the implementation of the above method embodiments can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless network, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0132] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the appended claims.

[0134] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for training an excavator crushing condition recognition model, characterized in that: include: Acquire training data; wherein the training data includes first sample features correspondingly extracted based on each first historical operating condition data, second sample features correspondingly extracted based on each second historical operating condition data, and third sample features correspondingly extracted based on each third historical operating condition data; the first historical operating condition data includes a plurality of pilot pressures collected from a first sample excavator with a crushing-only mode within a historical accumulated working time interval, the second historical operating condition data includes a plurality of dual-pump pressure sums collected from a second sample excavator of a dual-pump confluence type within a historical accumulated working time interval, and the third historical operating condition data includes a plurality of dual-pump pressure differences collected from a third sample excavator of a non-dual-pump confluence type within a historical accumulated working time interval; Based on the training data and the working condition labels corresponding to each sample feature in the training data, the machine learning model is trained to obtain a crushing working condition recognition model; wherein the working condition label is used to mark whether the sample excavator is in a crushing working condition.

2. The method according to claim 1, characterized in that The first sample characteristics include a pressure mean value, a pressure change rate, and a pressure change amplitude of the pilot pressure.

3. The method according to claim 1, characterized in that The second sample characteristics include a pressure sum mean, a pressure sum standard deviation, and a pressure sum change rate of the dual-pump pressure sum.

4. The method according to claim 1, characterized in that The third sample feature includes a pressure difference mean, a pressure difference standard deviation, and a pressure difference change rate of the dual-pump pressure difference.

5. The method according to any one of claims 1 to 4, characterized in that: The sample features in the training data also include frequency domain features, and the frequency domain features are obtained by the following method: The historical operating condition data is converted into frequency domain data, and frequency domain features are extracted based on the frequency domain data.

6. A method for identifying crushing conditions of an excavator, characterized in that: include: Acquire current working condition data of the target excavator to be identified; wherein the current working condition data includes a plurality of pilot pressures and a plurality of dual pump pressures and a plurality of dual pump pressure differences collected within a current accumulated working time interval; Extracting a first identification feature based on the pilot pressure in the current operating condition data, extracting a second identification feature based on the dual pump pressure in the current operating condition data, and extracting a third identification feature based on the dual pump pressure difference in the current operating condition data; Input the first identification feature, the second identification feature and the third identification feature into a crushing condition identification model respectively, and obtain corresponding identification results respectively; wherein the crushing condition identification model is obtained by training the excavator crushing condition identification model training method according to any one of claims 1 to 5; If there is at least one identification result indicating that the target excavator is in a crushing working condition, it is determined that the target excavator is in a crushing working condition.

7. An excavator crushing condition identification model training device, characterized in that: include: A training data acquisition module, used to acquire training data; wherein the training data includes first sample features correspondingly extracted based on each first historical operating condition data, second sample features correspondingly extracted based on each second historical operating condition data, and third sample features correspondingly extracted based on each third historical operating condition data; the first historical operating condition data includes a plurality of pilot pressures collected by a first sample excavator with a crushing-only mode within a historical accumulated working time interval, the second historical operating condition data includes a plurality of dual-pump pressure sums collected by a second sample excavator of a dual-pump confluence type within a historical accumulated working time interval, and the third historical operating condition data includes a plurality of dual-pump pressure differences collected by a third sample excavator of a non-dual-pump confluence type within a historical accumulated working time interval; The recognition model training module is used to train the machine learning model based on the training data and the working condition labels corresponding to each sample feature in the training data to obtain a crushing working condition recognition model; wherein the working condition label is used to mark whether the sample excavator is in a crushing working condition.

8. An excavator crushing condition identification device, characterized in that: include: An identification data acquisition module is used to obtain the current working condition data of the target excavator to be identified; wherein the current working condition data includes a plurality of pilot pressures and a plurality of dual pump pressures and a plurality of dual pump pressure differences collected within a current accumulated working time interval; an identification feature extraction module, configured to extract a first identification feature based on the pilot pressure in the current operating condition data, extract a second identification feature based on the dual pump pressure in the current operating condition data, and extract a third identification feature based on the dual pump pressure difference in the current operating condition data; A crushing condition identification module is used to input the first identification feature, the second identification feature and the third identification feature into a crushing condition identification model respectively, and obtain corresponding identification results respectively; wherein the crushing condition identification model is obtained by training according to the excavator crushing condition identification model training method according to any one of claims 1 to 5; if there is at least one identification result characterizing that the target excavator is in a crushing condition, it is determined that the target excavator is in a crushing condition.

9. An electronic device, characterized in that: comprising a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

10. An excavator, characterized in that: The excavator includes the electronic device according to claim 9.

Images