Earth-moving machine working condition identification method and device, storage medium and processor

Abstract

The application relates to the technical field of engineering machinery, and discloses a kind of earth-moving machinery working condition identification method, device, storage medium and processor.The identification method of the application comprises: obtaining the basic data of earth-moving machinery;According to the basic data, the working condition stage of the earth-moving machinery is identified using a pre-set working condition stage identification model, and a working condition stage identification result is obtained;According to the working condition stage identification result, the working condition type of the earth-moving machinery is identified using a pre-set working condition type identification model, and a working condition type identification result is obtained.A multilevel working condition identification method is realized, the complexity and computational power requirements of the working condition identification model are reduced, which is conducive to real vehicle deployment, improves the accuracy and stability of the identification result.Meanwhile, the working condition stage identification model and the working condition type identification model can cover all working condition stages and working condition types, widening the coverage of the identification model, which can be used in various scenarios, improving the environmental adaptability of the working condition identification method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering machinery, in particular, to a method for identifying working conditions of earth-moving machinery, an earth-moving machinery working condition identification device, a machine-readable storage medium and a processor. BACKGROUND

[0002] With the increase of labor and material costs in the project construction process and the improvement of quality and progress requirements, intelligent construction characterized by driving project decisions based on on-site equipment and personnel operation information has become the development direction in the future. In the construction process, excavators are responsible for completing 65-70% of the earthwork tasks, and the working condition changes, earthwork operation amount, fuel consumption and other operation information statistics of the excavators in the construction process are important contents of intelligent construction.

[0003] The most direct way to realize the operation information statistics of earth-moving machinery is to dispatch employees for on-site monitoring, but this method is high in cost and poor in reliability. Therefore, main engine manufacturers have developed various operation information statistics methods based on Internet of Things technology, in order to realize remote and accurate monitoring of the operation efficiency of excavators through the multi-source sensing information system carried by the excavators. However, these operation information statistics methods all need to identify the working conditions.

[0004] At present, for working condition identification, the working condition identification method based on visual information is often used in the prior art, and this working condition identification method has the shortcomings of narrow working condition type coverage and poor environmental adaptability. SUMMARY

[0005] The purpose of the present application is to provide a method for identifying working conditions of earth-moving machinery, an earth-moving machinery working condition identification device, a machine-readable storage medium and a processor. The method for identifying working conditions of earth-moving machinery can realize multi-level working condition identification, cover more working condition stages and working condition types, widen the coverage of the identification model, and improve the environmental adaptability of the working condition identification method.

[0006] To achieve the above purpose, the first aspect of the present application provides a method for identifying working conditions of earth-moving machinery, comprising:

[0007] obtaining basic data of the earth-moving machinery;

[0008] According to the basic data, a preset working condition stage identification model is used to identify the working condition stage of the earth-moving machinery, and a working condition stage identification result is obtained;

[0009] According to the working condition stage identification result, a preset working condition type identification model is used to identify the working condition type of the earth-moving machinery, and a working condition type identification result is obtained;

[0010] The data type of the basic data at least includes one of a main pump pressure signal, data collected by machine vision, a handle pilot control signal, and an actuator displacement signal.

[0011] In the embodiment of the present application, the construction process of the preset working condition stage recognition model includes:

[0012] Obtain first sample data, which includes main pump pressure data of each working condition stage under each working condition type, and working condition stage labels corresponding to the main pump pressure data;

[0013] Input the main pump pressure data of each working condition stage into a first neural network respectively to obtain predicted working condition stages;

[0014] Adjust parameters of the first neural network according to the predicted working condition stages and the working condition stage labels corresponding to the main pump pressure data in the first sample data to obtain a working condition stage recognition model.

[0015] In the embodiment of the present application, the construction process of the preset working condition type recognition model includes:

[0016] Obtain second sample data, which includes feature data of each working condition stage under each working condition type, and working condition type labels corresponding to the feature data;

[0017] Input the feature data of each working condition stage into a second neural network respectively to obtain predicted working condition types;

[0018] Adjust parameters of the second neural network according to the predicted working condition types and the working condition type labels corresponding to the feature data in the second sample data to obtain a working condition type recognition model.

[0019] In the embodiment of the present application, the basic data is a main pump pressure signal, and the main pump pressure signal includes a plurality of main pump pressure data within a first preset time range from a current time;

[0020] The working condition stage of the earth-moving machine is recognized according to the basic data by using a preset working condition stage recognition model to obtain a working condition stage recognition result, including:

[0021] A1: feature extraction is performed on the main pump pressure signal to obtain a feature vector;

[0022] A2: normalization processing is performed on the feature vector to obtain a normalized feature vector;

[0023] A3: the normalized feature vector is input into a preset working condition stage recognition model for working condition stage recognition to obtain a working condition stage recognition result;

[0024] A4: determining whether the working condition stage recognition process is ended;

[0025] A5: in the case where the working condition stage recognition process is ended, outputting a working condition stage recognition result;

[0026] A6: in the case where the working condition stage recognition process is not ended, removing the main pump pressure data farthest from the current time in the main pump pressure signal, and obtaining main pump pressure data at the next time, so as to update the main pump pressure signal, obtain a new main pump pressure signal, and return to execute A1.

[0027] In the embodiment of the present application, the working condition type of the earth-moving machine is identified by using a preset working condition type identification model according to the working condition stage recognition result, so as to obtain a working condition type identification result, which comprises:

[0028] B1: obtaining a plurality of working condition stage recognition results within a second preset time range from the current time, and grouping the plurality of working condition stage recognition results into a working condition stage vector;

[0029] B2: inputting the working condition stage vector into a preset working condition type identification model to identify the working condition type, so as to obtain a working condition type identification result;

[0030] B3: determining whether the working condition type identification process is ended;

[0031] B4: in the case where the working condition type identification process is ended, outputting a working condition type identification result;

[0032] B5: in the case where the working condition type identification process is not ended, removing the working condition stage recognition result farthest from the current time in the plurality of working condition stage recognition results, and obtaining a working condition stage recognition result at the next time, so as to update the working condition stage vector, obtain a new working condition stage vector, and return to execute B2.

[0033] In the embodiment of the present application, after obtaining the working condition type identification result, the following steps are further included:

[0034] According to the working condition stage recognition result and the working condition type identification result, statistical conditions of working condition information are preset, and the working condition information is counted according to the statistical conditions, so as to obtain a working condition information statistical result;

[0035] Obtaining fuel consumption rate information;

[0036] According to the working condition stage recognition result, the working condition type identification result, the working condition information statistical result and the fuel consumption rate information, the working information of the earth-moving machine is counted in the time dimension, so as to obtain a first working information statistical result.

[0037] In the embodiment of the present application, the following steps are further included:

[0038] acquire topographic information of a working range of the earth-moving machine, and form an electronic map of a working environment according to the topographic information;

[0039] acquire a positioning signal of the earth-moving machine and motion information of an actuator in real time;

[0040] determine a position of the earth-moving machine in the electronic map of the working environment according to the positioning signal of the earth-moving machine, and obtain a working device track of the earth-moving machine according to the motion information of the actuator of the earth-moving machine;

[0041] determine a worked earth volume based on the electronic map of the working environment and the working device track of the earth-moving machine;

[0042] The working device track is a track formed by the motion information of the actuator of the earth-moving machine changing over time.

[0043] In the embodiments of the present application, the determination of the worked earth volume based on the electronic map of the working environment and the working device track of the earth-moving machine comprises:

[0044] divide the working range in the electronic map of the working environment into a plurality of uniform cubes;

[0045] determine an intersection state of each cube with the working device track respectively to obtain a plurality of cube states;

[0046] obtain the worked earth volume according to the cube states and volumes of the cubes.

[0047] In the embodiments of the present application, at least one of the following steps is further included:

[0048] calculate an earth-moving efficiency according to the worked earth volume and a total working duration of earth excavation, and the first working information statistical result at least includes the total working duration of earth excavation;

[0049] calculate a single earth excavation working volume according to the worked earth volume and a cycle number of earth excavation, and the working condition information statistical result at least includes the cycle number of earth excavation.

[0050] The second aspect of the present application provides an earth-moving machine working condition recognition device, comprising:

[0051] an acquisition module configured to acquire basic data of the earth-moving machine; wherein the data types of the basic data at least include one of a main pump pressure signal, data collected by machine vision, an operation handle pilot control signal, and an actuator displacement signal;

[0052] The first identification module is configured to identify the working condition stage of the earth-moving machine according to the basic data by using a preset working condition stage identification model, and obtain a working condition stage identification result.

[0053] The second identification module is configured to identify the working condition type of the earth-moving machine according to the working condition stage identification result by using a preset working condition type identification model, and obtain a working condition type identification result.

[0054] The third aspect of the present application provides a processor configured to execute the earth-moving machine working condition identification method.

[0055] The fourth aspect of the present application provides a machine-readable storage medium having instructions stored thereon, which, when executed by a processor, cause the processor to be configured to execute the earth-moving machine working condition identification method.

[0056] According to the above technical solution, based on the basic data of the earth-moving machine, the working condition stage identification model and the working condition type identification model are used to sequentially perform working condition stage identification and working condition type identification, so as to obtain the working condition stage identification result and the working condition type identification result. Based on the basic data of the earth-moving machine, the working condition identification of the earth-moving machine is completed in a multi-step manner of working condition stage-working condition type, so as to realize a multi-level working condition identification method. Compared with a single-step working condition identification method, the complexity and the power requirement of the working condition identification model can be reduced, which is beneficial for real vehicle loading, and the accuracy and stability of the identification result are improved. At the same time, the working condition stage identification model and the working condition type identification model can cover more working condition stages and working condition types, thereby widening the coverage of the identification model, so as to be used in various scenarios, and the environmental adaptability of the working condition identification method is improved.

[0057] Other features and advantages of the present application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0058] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, and are used together with the following specific embodiments to explain the present application, but do not constitute a limitation on the present application. In the drawings:

[0059] Figure 1 The flowchart of the earth-moving machine working condition identification method according to the embodiment of the present application is schematically shown;

[0060] Figure 2 The multi-level multi-dimensional hydraulic excavator operation information statistical flowchart according to the embodiment of the present application is schematically shown;

[0061] Figure 3 The backhoe excavator structure simplified diagram according to the embodiment of the present application is schematically shown;

[0062] Figure 4 The D-H coordinate system of the excavator working device is schematically shown according to the embodiment of the application;

[0063] Figure 5 The structural block diagram of the earth-moving machine working condition recognition device is schematically shown according to the embodiment of the application;

[0064] Figure 6 The internal structure diagram of the computer device is schematically shown according to the embodiment of the application.

[0065] Explanation of reference signs

[0066] 1 - lower car; 2 - upper car; 3 - boom; 4 - boom hydraulic cylinder; 5 - stick hydraulic cylinder; 6 - stick; 7 - bucket hydraulic cylinder; 8 - rocker arm; 9 - connecting rod; 10 - bucket; 410 - acquisition module; 420 - first recognition module; 430 - second recognition module; A01 - processor; A02 - network interface; A03 - internal memory; A04 - display screen; A05 - input device; A06 - non-volatile storage medium; B01 - operating system; B02 - computer program. DETAILED DESCRIPTION

[0067] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. It should be understood that the specific embodiments described herein are only used to explain and illustrate the embodiments of the present application and should not be used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0068] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative positional relationship, movement condition, etc. between components in a certain specific posture (as shown in the drawings), and if the specific posture changes, the directional indications also change accordingly.

[0069] In addition, if the description of "first", "second" and the like is involved in the embodiments of the present application, the description of "first", "second" and the like is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the technical features indicated or the number of technical features indicated. Therefore, the features limited by "first", "second" can be explicitly or implicitly included at least one of the features. In addition, the technical solutions of various embodiments can be combined with each other, but it must be based on the realization of ordinary skilled in the art, when the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, nor in the protection scope claimed by the present application.

[0070] Please refer to Figure 1 and Figure 2 , Figure 1 The flowchart of the working condition recognition method of the earth-moving machine according to the embodiments of the present application is schematically shown, Figure 2 The multi-level and multi-dimensional hydraulic excavator operation information statistical flowchart according to the embodiments of the present application is schematically shown.

[0071] As Figure 1 shown, in an embodiment of the present application, a working condition recognition method of an earth-moving machine is provided. It should be noted that the working condition recognition method of the earth-moving machine provided in the embodiment can be applied to earth-moving machines such as excavators, loaders, pushers or graders. In order to facilitate the description of the scheme, the earth-moving machine in the embodiment is mainly taken as an excavator for illustration. The working condition recognition method of the earth-moving machine comprises the following steps:

[0072] Step 210: acquiring the basic data of the earth-moving machine; wherein the data type of the basic data at least includes one of the main pump pressure signal, the data collected by machine vision, the operation handle pilot control signal and the actuator displacement signal;

[0073] Step 220: according to the basic data, using a pre-set working condition stage recognition model to recognize the working condition stage of the earth-moving machine, and obtaining a working condition stage recognition result;

[0074] Step 230: according to the working condition stage recognition result, using a pre-set working condition type recognition model to recognize the working condition type of the earth-moving machine, and obtaining a working condition type recognition result.

[0075] By the technical solution, the basic data of the earth-moving machine is first acquired, and then based on the basic data, the working condition stage recognition model and the working condition type recognition model are used to sequentially perform working condition stage recognition and working condition type recognition, so as to obtain the working condition stage recognition result and the working condition type recognition result. Based on the basic data of the earth-moving machine, the working condition recognition of the earth-moving machine is completed in a multi-step manner of working condition stage-working condition type, so as to realize a multi-level working condition recognition method. Compared with a single-step working condition recognition method, the complexity and the computing power requirement of the working condition recognition model are reduced, which is beneficial to real vehicle loading, and the accuracy and stability of the recognition result are improved. Meanwhile, the working condition stage recognition model and the working condition type recognition model can cover more working condition stages and working condition types, which widens the coverage of the recognition model, so that the working condition recognition method can be used in various scenes, and the environmental adaptability of the working condition recognition method is improved.

[0076] In the implementation process, the working condition stage recognition result and the working condition type recognition result are respectively recognized by using the working condition stage recognition model and the working condition type recognition model, and the output of the working condition stage recognition model is used as the input of the working condition type recognition model, so as to realize a multi-level working condition recognition method, reduce the calculation amount, and facilitate rapid obtaining of the working condition recognition result.

[0077] In this embodiment, the basic data of the earth-moving machine is a data source for working condition recognition, which can be at least one of a main pump pressure signal, data collected by machine vision, an operation handle pilot control signal, an actuator displacement signal, and the like.

[0078] It should be noted that, in order to facilitate the description of the scheme, the main pump pressure signal is mainly used as the data source for working condition recognition in this embodiment, that is, the basic data is the main pump pressure signal. The main pump pressure signal of the earth-moving machine can be collected by a pressure sensor on the earth-moving machine. The main pump pressure signal can be main pump pressure data collected at the current time, or main pump pressure data within a certain time range. For example, in the case of an excavator, the pressure sensor is used to collect the main pump pressure signal of the excavator within 0.5s from the current time. For different types of earth-moving machines, the main pump pressure signal can be one or more groups, for example, for a small excavator, only one group of main pump pressure signal is included, and for a large or medium-sized excavator, two groups of main pump pressure signal are included.

[0079] Correspondingly, step 220 can use a preset working condition stage recognition model to recognize the working condition stage of the earth-moving machine according to the main pump pressure signal, to obtain a working condition stage recognition result.

[0080] In the case of the earth-moving machine being an excavator, the working condition types of the excavator can include earth excavation, grading, slope repairing, crushing, whole vehicle walking, and idling; wherein the earth excavation working condition type of the excavator includes five working condition stages of excavation preparation, excavation, lifting and rotating, unloading, and empty bucket returning; the grading working condition type includes two working condition stages of bucket and arm outward swinging and bucket and arm inward swinging; the slope repairing working condition type includes two working condition stages of boom lifting and arm inward swinging and boom descending and arm outward swinging; the crushing working condition type includes two working condition stages of crushing impact and crushing point adjustment; the whole vehicle walking and idling are both working condition types and working condition stages, and are not subdivided. The working condition types and working condition stages of other earth-moving machines can be determined according to actual conditions, and will not be described here.

[0081] The working condition stage recognition model can be pre-installed in the earth-moving machine. In some embodiments, the construction process of the pre-installed working condition stage recognition model includes the following steps:

[0082] First, first sample data is obtained, which includes main pump pressure data of each working condition stage under each working condition type, and a working condition stage label corresponding to the main pump pressure data; in this embodiment, the first sample data contains multiple groups of data, and each group of data includes a main pump pressure data and a working condition stage label corresponding to the main pump pressure data. The division of each working condition stage can be based on the main pump pressure waveform corresponding to each working condition stage as a working condition stage division mark. For example: in the case of the earth-moving machine being an excavator, since the main pump pressure signals under different working condition stages are different, the main pump pressure waveform corresponding to each working condition stage under different working condition types can be used as a working condition stage division mark to segment the working cycle of the excavator, and the segmentation result is the corresponding working condition stage. Taking the earth excavation working condition type as an example, after the working cycle is segmented, there are five working condition stages of excavation preparation, excavation, lifting and rotating, unloading, and empty bucket returning. It should be noted that the main pump pressure waveform here refers to the main pump pressure signal.

[0083] Then, the main pump pressure data of each working condition stage is input into the first neural network to obtain a predicted working condition stage;

[0084] Then, the predicted working condition stage is compared with the working condition stage label corresponding to the main pump pressure data in the first sample data to obtain a working condition stage comparison result;

[0085] Finally, the parameters of the first neural network are adjusted according to the working condition stage comparison result to obtain a working condition stage recognition model.

[0086] In the embodiment, the first neural network can be a linear neural network, a feedback neural network, a multi-layer feedforward neural network (BP neural network), etc., wherein the BP neural network has strong non-linear mapping capability and flexible network structure. The number of intermediate layers of the network and the number of neurons of each layer can be arbitrarily set according to specific conditions, and the network has strong generalization capability and fault tolerance. A more stable and reliable working condition stage recognition model can be obtained by using the BP neural network. In the case of the earth-moving machine being a excavator, a working condition stage recognition model of the excavator can be established based on the BP neural network, the main pump pressure data of each working condition stage under each working condition type is taken as the input of the model, the predicted working condition stage is obtained, the predicted working condition stage and the working condition stage label corresponding to the main pump pressure data are input into a preset loss function, a corresponding loss value is obtained, and the model parameters are adjusted according to the loss value, so that the predicted working condition stage is the same as the working condition stage label corresponding to the main pump pressure data, so that the model has sufficient recognition accuracy, and finally a working condition stage recognition model is trained.

[0087] Correspondingly, a working condition type recognition model can also be constructed in advance, and the construction process of the preset working condition type recognition model includes the following steps:

[0088] First, second sample data is obtained, the second sample data includes feature data of each working condition stage under each working condition type, and a working condition type label corresponding to the feature data; in the embodiment, the second sample data contains multiple groups of data, each group of data includes feature data of a working condition stage, and a working condition type label corresponding to the feature data. The division of each working condition type can be based on the waveform feature corresponding to each working condition stage as the working condition type division mark. For example: in the case of the earth-moving machine being a excavator, the waveform feature of the working condition stage corresponding to the working condition type under the working condition type is taken as the recognition mark of the working condition type, the working type of the excavator is divided, and the division result is the working condition type corresponding to the current time. Taking the earth-moving working condition type as an example, when the working condition stage at the current time is any one of the digging preparation, digging, lifting and rotating, unloading and emptying return, the working condition type at the current time is earth-moving.

[0089] Then, the feature data of each working condition stage is input into the second neural network respectively to obtain a predicted working condition type;

[0090] Then, the predicted working condition type is compared with the working condition type label corresponding to the feature data in the second sample data to obtain a working condition type comparison result;

[0091] Finally, the parameters of the second neural network are adjusted according to the working condition type comparison result to obtain a working condition type recognition model.

[0092] In the embodiment, the second neural network can be a linear neural network, a feedback neural network, a multi-layer feedforward neural network (BP neural network), etc., wherein the BP neural network has strong non-linear mapping capability and flexible network structure. The number of intermediate layers of the network and the number of neurons of each layer can be arbitrarily set according to specific conditions, and has strong generalization capability and fault tolerance. Using the BP neural network can obtain a more stable and reliable working condition type recognition model. In the case of the earth-moving machine being a excavator, a working condition type recognition model of the excavator can be established based on the BP neural network, the waveform features of the working condition stage under all working condition types are taken as the model input, the predicted working condition type is obtained, the predicted working condition type and the working condition type label corresponding to the feature data are input into a preset loss function, a corresponding loss value is obtained, and the model parameters are adjusted according to the loss value, so that the predicted working condition type is the same as the working condition type label corresponding to the feature data, so that the model has sufficient recognition accuracy, and finally the working condition type recognition model is trained.

[0093] In the above implementation process, the working condition stage recognition model and the working condition type recognition model are trained by using the neural network, which helps to improve the reliability of the working condition recognition result. The BP neural network model is selected, which can reduce the cost of working condition recognition and is more convenient for mounting on the engineering machinery.

[0094] In the above implementation process, the working condition stage recognition model and the working condition type recognition model are trained by using the neural network, which helps to improve the reliability of the working condition recognition result. The BP neural network model is selected, which can reduce the cost of working condition recognition and is more convenient for mounting on the engineering machinery.

[0095] In one embodiment, the working condition stage result can be obtained in real time, and the working condition stage of the earth-moving machine is identified by using the preset working condition stage recognition model according to the basic data to obtain a working condition stage recognition result, including the following steps:

[0096] Step A1: feature extraction is performed on the main pump pressure signal to obtain a feature vector; in this embodiment, the feature extraction includes reducing the noise and transient interference of the main pump pressure signal by using mean filtering; then, according to the main controller computing power, the system sampling method is used to reduce the frequency of the main pump pressure signal; finally, the time domain characteristic value of the reduced frequency main pump pressure signal is extracted to obtain the feature vector. The above-mentioned extraction of the time domain characteristic value of the reduced frequency main pump pressure signal can be obtained by calculating the mean value and variance of the main pump pressure signal. By using system sampling to reduce the data sampling frequency, the low-cost and low-computing power requirement of working condition recognition is ensured, which is more convenient for loading to earth-moving machinery.

[0097] Taking an excavator as an example, for a small excavator with only one main pump, the feature vector is composed of the mean value and variance of the No. 1 main pump pressure signal, which can be represented as: X = [x1, x2], wherein X is the feature vector constructed by the time domain characteristic values; x1 is the mean value of the No. 1 main pump pressure signal; x2 is the variance of the No. 1 main pump pressure signal; for a large and medium-sized excavator with two main pumps, the feature vector is composed of the mean value and variance of the No. 1 main pump pressure signal, the mean value and variance of the No. 2 main pump pressure signal, and the mean value and variance of the difference between the No. 1 and No. 2 main pump pressure signals, which can be represented as: X = [x1, x2, x3, x4, x5, x6], wherein X is the feature vector constructed by the time domain characteristic values; x1 is the mean value of the No. 1 main pump pressure signal; x2 is the variance of the No. 1 main pump pressure signal; x3 is the mean value of the No. 2 main pump pressure signal; x4 is the variance of the No. 2 main pump pressure signal; x5 is the mean value of the difference between the No. 1 and No. 2 main pump pressure signals; x6 is the variance of the difference between the No. 1 and No. 2 main pump pressure signals.

[0098] Step A2: normalizing the feature vector to obtain a normalized feature vector; in this embodiment, the above-mentioned normalization can be to calculate the normalized feature values by substituting the feature values in the feature vector into the normalization formula, thereby obtaining the normalized feature vector. The normalization formula can be:

[0099] wherein x new is the normalized feature value; x is the feature value before normalization; x max is the maximum value of the feature value of the corresponding type in the feature vector; x min is the minimum value of the feature value of the corresponding type in the feature vector. In actual application, other normalization methods such as nonlinear normalization can also be used.

[0100] Step A3: inputting the normalized feature vector into a pre-set working condition stage recognition model to perform working condition stage recognition and obtaining a working condition stage recognition result;

[0101] In the case of the earth-moving machine being an excavator, the normalized feature vector is input to the working condition stage recognition model, and the working condition stage of the excavator can be determined according to the sequence number of the maximum output probability value in the output result of the working condition stage recognition model. For example, when the sequence number of the maximum probability value is 1-5, the working condition stage of the excavator at the current moment is the digging preparation, digging, lifting and rotating, unloading, and emptying return stage under the earth-moving working condition type; when the sequence number of the maximum probability value is 6-7, the working condition stage of the excavator at the current moment is the bucket and arm outside swing and the bucket and arm inside swing stage under the land leveling working condition type; when the sequence number of the maximum probability value is 8-9, the working condition stage of the excavator at the current moment is the boom lifting and arm inside swing and the boom descending and arm outside swing stage under the slope repairing working condition type; when the sequence number of the maximum probability value is 10-11, the working condition stage of the excavator at the current moment is the breaking impact and breaking point adjustment stage under the breaking working condition type; and when the sequence number of the maximum probability value is 12-13, the working condition stage of the excavator at the current moment is the whole vehicle walking and idling machine stage. It should be noted that the above sequence number can be set when the working condition stage recognition model is pre-trained, and the sequence number of the maximum output probability value is set as the output result.

[0102] In an embodiment, the obtained working condition stage recognition result can be stored in a hardware device such as a processor, a cache, etc.

[0103] In an embodiment, the obtained working condition stage recognition result can be output to an output device in real time, and the output device can be a memory, a display device, a terminal, a communication module, etc., so that the user can obtain the working condition stage recognition result in time. For example, the working condition stage recognition result stored in the processor can be output to the display device.

[0104] In an embodiment, in order to further improve the accuracy of working condition stage recognition, the working condition stage recognition can be performed in a loop until an end signal is received, and the working condition stage recognition result is output. Specifically, after step A3 is performed, step A4 can be further performed: determining whether the working condition stage recognition process is ended. In this embodiment, whether the working condition stage recognition process is ended can be determined by determining whether an end signal is received. The end signal can be obtained by starting a switch by a worker according to the actual situation. The working condition stage recognition process can also be determined by determining whether the power is off.

[0105] Step A5: outputting the working condition stage recognition result when the working condition stage recognition process is ended.

[0106] In the embodiment, the obtained working condition stage recognition result can be output from the hardware device to an output device, which can be a memory, a display device, a terminal, a communication module, etc., so that the user can obtain the working condition stage recognition result in time. For example, the working condition stage recognition result stored in the processor can be output to the display device.

[0107] Step A6: In the case where the working condition stage recognition process is not completed, the main pump pressure data farthest from the current time in the main pump pressure signal is removed, and the main pump pressure data at the next time is obtained to update the main pump pressure signal, so as to obtain a new main pump pressure signal, and A1-A4 is executed.

[0108] In the embodiment, since the main pump pressure signal comprises the main pump pressure data in a certain time range, in the case where the working condition stage recognition process is not completed, the main pump pressure data farthest from the current time can be removed, and the main pump pressure data at the next time is added, so as to form a new main pump pressure signal, and then steps A1-A4 are repeated until the recognition is completed, so as to ensure the accuracy and real-time performance of the working condition stage recognition result.

[0109] Correspondingly, in order to make the obtained working condition type recognition result more accurate, the working condition stage recognition result in a certain time range can be used as a data source to identify the working condition type,

[0110] In an embodiment, the working condition type recognition result is a real-time output working condition type recognition result. The working condition type recognition result is obtained by using a preset working condition type recognition model to identify the working condition type of the earth-moving machine according to the working condition stage recognition result, and comprises:

[0111] Step B1: Obtain a plurality of working condition stage recognition results within a second preset time range from the current time, and combine the plurality of working condition stage recognition results to form a working condition stage vector. For example, in the case of an excavator, the recognition result within 0.5 s from the current time output by the excavator working condition stage recognition model is received, and the working condition stage recognition results are combined to form a vector, so as to obtain a working condition stage vector.

[0112] It should be noted that the first preset time and the second preset time in the embodiment can be the same or different, and are set according to actual needs.

[0113] Step B2: inputting the working condition stage vector into a preset working condition type recognition model to identify the working condition type, so as to obtain the working condition type recognition result;

[0114] In the case of the earth-moving machine being an excavator, the working condition stage vector is input to the working condition type identification model, and the working condition type of the excavator is determined according to the serial number of the maximum output probability value in the model output result. For example, when the serial number of the maximum probability value is 1, the working condition type of the excavator at the current time is earth-moving; when the serial number of the maximum probability value is 2, the working condition type of the excavator at the current time is grading; when the serial number of the maximum probability value is 3, the working condition type of the excavator at the current time is slope repairing; when the serial number of the maximum probability value is 4, the working condition type of the excavator at the current time is crushing; and when the serial number of the maximum probability value is 5-6, the working condition types of the excavator at the current time are vehicle walking and engine idling, respectively. It should be noted that the above serial numbers can be set when the working condition stage identification model is pre-trained, and the serial number of the maximum output probability value is set as the output result.

[0115] In an embodiment, the obtained working condition type identification result can be stored in a hardware device such as a processor, a cache, etc.

[0116] In an embodiment, the obtained working condition type identification result can be output from the hardware device to an output device in real time, and the output device can be a memory, a display device, a terminal, a communication module, etc., so that the user can obtain the working condition type identification result in time. For example, the working condition type identification result stored in the processor can be output to the display device.

[0117] In an embodiment, in order to further improve the accuracy of working condition type identification, the working condition type identification can be performed in a loop until an end signal is received, and the working condition type identification result is output. Specifically, after step B2 is performed, step B3 of determining whether the working condition type identification process is ended can be performed. In this embodiment, the determination of whether the working condition type identification process is ended is similar to the determination of whether the working condition stage identification process is ended in step A4, which is determined by judging whether an end signal is received. Therefore, the details are not repeated here.

[0118] Step B4: In the case of the working condition type identification process being ended, the working condition type identification result is output.

[0119] In this embodiment, the obtained working condition type identification result can be output from the hardware device to an output device, and the output device can be a memory, a display device, a terminal, a communication module, etc., so that the user can obtain the working condition type identification result in time. For example, the working condition type identification result stored in the processor can be output to the display device.

[0120] Step B5: In the case where the working condition type identification process has not ended, the working condition stage identification result farthest from the current time among the plurality of working condition stage identification results is eliminated, the working condition stage identification result at the next time is obtained, the working condition stage vector is updated to obtain a new working condition stage vector, and B2-B3 are executed.

[0121] In the embodiment, since the working condition stage vector is composed of working condition stage identification results at a plurality of time points, in the case where the working condition type identification process has not ended, the working condition stage identification result farthest from the current time can be eliminated, and then the working condition stage identification result at the next time is added to form a new working condition stage vector. Then, steps B2-B3 are repeated until the identification ends, so as to ensure the accuracy and real-time performance of the working condition type identification result.

[0122] In the implementation process, the main pump pressure signal of the earth-moving machine is obtained; then, according to the main pump pressure signal, a preset working condition stage identification model is used to identify the working condition stage of the earth-moving machine to obtain a working condition stage identification result; and according to the working condition stage identification result, a preset working condition type identification model is used to identify the working condition type of the earth-moving machine to obtain a working condition type identification result. Based on the main pump pressure signal, the working condition stage identification model and the working condition type identification model are used for identification in sequence, so as to obtain the working condition stage identification result and the working condition type identification result. The main pump pressure generated by the basic action of the actuator is taken as a data source, and the working condition identification of the earth-moving machine is completed in a multi-step manner of working condition stage-working condition type, so as to realize a multi-level working condition identification method. Compared with a single-step working condition identification method, the complexity and the power requirement of the working condition identification model are reduced, which is beneficial to real vehicle loading, and the accuracy and stability of the identification result are improved. Meanwhile, the working condition stage identification model and the working condition type identification model can cover more working condition stages and working condition types, so as to widen the coverage of the identification model, thereby being applicable to various scenes and improving the environmental adaptability of the working condition identification method.

[0123] By dividing the working process of the excavator into six working condition types of earth excavation, land leveling, slope repairing, crushing, whole vehicle walking and machine idling, and further defining the working condition stages covered by each working condition type, the mapping relationship between the basic action of the actuator and the main pump pressure waveform under each working condition stage is used to realize the identification of the working condition stage and the working condition type from bottom to top. This multi-level identification method can effectively improve the reliability of the model. Moreover, the main pump pressure signal is used as the original signal source, and the BP neural network is used to establish the identification model, so as to reduce the cost and the power requirement, and make the model more suitable for the main machine configuration of the engineering machinery.

[0124] After obtaining the working condition type identification result, working condition information statistics and time-dimension working information statistics can be performed, which specifically include the following steps:

[0125] Firstly, according to the working condition stage recognition result and the working condition type recognition result, the working condition information is counted according to the preset working condition information counting condition to obtain a working condition information counting result. In this embodiment, the working condition information counting result includes the complete cycle counting of each working condition stage and each working condition type. The counting process can be counting by using the working condition information counting condition. In specific implementation, the above counting can be completed by using the conditional form of IF-AND-THEN. The working condition stage recognition result and the working condition type recognition result are matched with the IF part one by one. If the IF part of a certain condition is met, the THEN part is executed to change the counting result. Otherwise, the original counting result remains unchanged.

[0126] In the case of the earth-moving machine being an excavator, the above IF-AND-THEN condition can include:

[0127] Condition one: IF the working condition stage result at the current moment is different from the working condition stage result at the last moment, THEN the cycle counting of the working condition stage result corresponding to the last moment is increased by 1;

[0128] Condition two: IF the working condition stage result at the last moment is any one of the bucket return, the bucket and arm retraction, the boom descending arm, the broken point adjustment, the whole vehicle walking, and the engine idling, AND the working condition stage result at the current moment is different from the working condition stage result at the last moment, THEN the cycle counting of the working condition type result corresponding to the working condition stage result at the last moment is increased by 1. It should be noted that the above condition can be set according to different working condition types.

[0129] By counting according to the working condition stage recognition result and the working condition type recognition result by using the working condition information counting condition, the working condition information can be quickly and effectively counted, which is helpful for the project manager to make reasonable project decisions.

[0130] Then, the fuel consumption rate information is obtained. In this embodiment, the fuel consumption rate information can be CAN bus fuel consumption rate information, which can be obtained from the control module of the earth-moving machine.

[0131] Finally, according to the working condition stage recognition result, the working condition type recognition result, the working condition information counting result, and the fuel consumption rate information, the working information of the earth-moving machine is counted in the time dimension to obtain a first working information counting result. In this embodiment, the working information counting in the time dimension includes the total time and the time proportion of each working condition stage, the average fuel consumption rate and the fuel consumption rate proportion of each working condition stage, the total time, the single cycle duration and the time proportion of each working condition type, and the average fuel consumption rate and the fuel consumption rate proportion of each working condition type.

[0132] In the case of the earth-moving machine being a shovel, the operation information statistics of the time dimension can be calculated by the following formula:

[0133]

[0134] where Δt is the working condition stage recognition time interval; M1 i is the number of occurrences of each working condition stage in the working condition stage recognition result; T1 i is the total duration of each working condition stage; η1 i is the time proportion of each working condition stage; o1 i is the instantaneous fuel consumption rate of each working condition stage; O1 i is the average fuel consumption rate of each working condition stage; η2 i is the average fuel consumption rate proportion of each working condition stage; M1 i , T1 i , η1 i , o1 i , O1 i , η2 i , i = 1, 2, …, 13, respectively represent the digging preparation, digging, lifting and rotating, unloading, empty bucket return, bucket and arm outward swinging, bucket and arm inward swinging, boom lifting and arm inward swinging, boom lowering and arm outward swinging, breaking impact, breaking point adjustment, whole vehicle walking, and idling working condition stages; M2 i is the number of occurrences of each working condition type in the working condition type recognition result; N i is the cycle number of each working condition type, and is the multi-level working condition information statistics result; T2 i is the total duration of each working condition type; t i is the single cycle duration of each working condition type; η3 i is the single cycle duration proportion of each working condition type; o2 i is the instantaneous fuel consumption rate of each working condition type; O2 i is the average fuel consumption rate of each working condition type; η4 i is the average fuel consumption rate proportion of each working condition type; M2 i , N i , T2 i , t i , η3 i , o2 i , O2 i , η4 i , i = 1, 2, …, 6, respectively represent the earth digging preparation, land leveling, slope repairing, breaking, whole vehicle walking, and idling working condition types.

[0135] In the implementation process, the multi-level excavator working condition information statistics and the time dimension operation information statistics are completed according to the working condition stage, the working condition type identification result and the CAN bus fuel consumption rate information, the statistics covers the information of working conditions and energy consumption, and the details of the working condition type and the working condition stage in the excavator operation process are revealed, which is beneficial to the project manager to make reasonable project decisions.

[0136] At present, the operation amount of the entire construction site is measured in the construction process to determine the overall progress of the project. The way is to send workers to watch, or some simple equipment is used to measure the operation amount of the entire construction site, or the difference of the electronic map in a period of time is determined by regular unmanned aerial vehicle cruising, and the overall project construction earthwork quantity is determined. However, these methods cannot measure the operation earthwork quantity of each earthmoving machine, and the operation efficiency of a single earthmoving machine cannot be measured.

[0137] In the embodiment, the operation information statistics can also be performed from the spatial dimension, including statistics of the operation earthwork quantity of a single earthmoving machine, specifically including the following steps:

[0138] Firstly, the topographic information of the earthmoving machine to be operated is obtained, and an operation environment electronic map is formed according to the topographic information; in the embodiment, the topographic information can be 3D topographic information, taking an excavator as an example, the way of obtaining 3D topographic information of the excavator operation range includes single artificial survey, unmanned aerial vehicle laser radar technology, machine vision technology, millimeter wave radar technology or multi-mode combination. The unmanned aerial vehicle laser radar completes the topographic point cloud model in the operation range of the excavator through the flight of the aircraft and the scanning of the laser pulse; the visual sensor is installed on the front and rear of the top of the excavator cab, and on the left and right sides of the vehicle body; the millimeter wave radar is installed on the front, rear, left and right of the lower vehicle body. All the above technologies are ultimately used to obtain the topographic elevation and slope information in the operation range of the excavator, and form the operation environment electronic map.

[0139] Then, the positioning signal and the actuator motion information of the earthmoving machine are obtained in real time; in the embodiment, taking the excavator as an example, the positioning signal of the excavator refers to the absolute position of the upper swing center, and the acquisition method includes single differential GPS high-precision positioning system, inertial navigation unit or multi-mode combination. The actuator motion information is obtained by installing displacement, inclination or position sensors on the actuator, obtaining the actuator driving space or joint space coordinates, and finally converting into the bucket tooth tip pose space coordinates by using the excavator kinematics model. Specifically, the following steps are included:

[0140] Firstly, the initial space coordinates of the actuator are obtained in real time; in the embodiment, the initial space coordinates can be actuator driving space or joint space coordinates.

[0141] Second, based on the excavator kinematic model, the initial spatial coordinates are converted into actuator pose spatial coordinates;

[0142] Third, according to the actuator pose spatial coordinates, actuator motion information is obtained.

[0143] Please refer to Figure 3 and Figure 4 , Figure 3 schematically shows a simplified structure of a backhoe excavator according to an embodiment of the present application, Figure 4 schematically shows a D-H coordinate system of an excavator working device according to an embodiment of the present application. In the case of an excavator, the excavator includes a lower vehicle 1, an upper vehicle 2, a boom 3, a boom hydraulic cylinder 4, a stick hydraulic cylinder 5, a stick 6, a bucket hydraulic cylinder 7, a swing arm 8, a connecting rod 9, and a bucket 10. Coordinate systems are established at the lower vehicle 1, the upper vehicle 2, the boom 3, the stick 6, and the bucket 10, respectively, and are denoted as O0, O1, O2, O3, and O4, wherein O0 is a base coordinate system, O1 is an upper vehicle coordinate system, O2 is a boom coordinate system, O3 is a stick coordinate system, and O4 is a bucket coordinate system. The positional relationship between adjacent coordinate systems is described using a offset, an angle, a rod length, and a torsion angle, which are defined as follows:

[0144] offset s i : distance along the z i axis from the x i axis to the x j axis, which is positive in the same direction as the z i axis;

[0145] angle θ i : angle along the z i axis from the x i axis to the x j axis, which is positive in the counterclockwise direction, corresponding to the joint space;

[0146] rod length h i : distance along the x j axis from the z i axis to the z j axis, which is positive in the same direction as the x j axis;

[0147] torsion angle α i : angle along the x j axis from the z i axis to the z j axis, which is positive in the counterclockwise direction.

[0148] excavator driving space: composed of a swing motor angle L0, a boom hydraulic cylinder length L1, a stick hydraulic cylinder length L2, and a bucket hydraulic cylinder length L3, denoted as [L0, L1, L2, L3]T ;

[0149] Excavator joint space: composed of the angle θ0 between the lower car 1 and the upper car 2, the angle θ1 between the upper car 2 and the boom 3, the angle θ2 between the boom 3 and the stick 6, and the angle θ3 between the stick 6 and the bucket 10, represented as [θ0, θ1, θ2, θ3] T ;

[0150] Excavator pose space: composed of the position of the bucket 10 in the base coordinate system and the pose angle (the angle of the stop surface to the connecting line of the hinge point of the bucket 10 and the tooth tip of the bucket 10), represented as [x, y, z, ζ] T .

[0151] The joint space coordinates of the excavator can be obtained by the formula of driving space to joint space, and then the pose space coordinates can be obtained by the formula of joint space to pose space.

[0152] Wherein, the formula of driving space to joint space is:

[0153]

[0154] The formula of joint space to pose space is:

[0155]

[0156] Wherein, ∠XYZ represents the angle between straight line XY and straight line YZ, L XY is the distance between the hinge points X and Y, and similar variables are the angle between two straight lines or the distance between two hinge points, which can be identified with reference to Figure 2 ; α is the angle between the connecting line of the hinge points A and F and the horizontal plane; θ0 is the angle between the lower car 1 and the upper car 2; θ1 is the angle between the upper car 2 and the boom 3; θ2 is the angle between the boom 3 and the stick 6; θ3 is the angle between the stick 6 and the bucket 10; h1 is the length of the boom 3 AB; h2 is the length of the stick 6 BG; h3 is the length of the bucket 10 GJ; the pose space coordinates are represented as: [x, y, z, ζ] T , wherein x, y, z are the positions of the actuator in the base coordinate system (such as the position of the bucket 10 in the base coordinate system); ζ is the pose angle, that is, the angle of the stop surface to the connecting line of the hinge point of the bucket 10 and the tooth tip of the bucket 10.

[0157] Then, according to the positioning signal of the earth-moving machine, the position of the earth-moving machine is determined in the electronic map of the working environment, and according to the motion information of the actuators of the earth-moving machine, the working device trajectory of the earth-moving machine is obtained; wherein the working device trajectory is a trajectory formed by the motion information of the actuators of the earth-moving machine changing with time. In this embodiment, as the motion information of the actuators changes over time, taking the excavator as an example, it is embodied as the position of the bucket 10 tooth tip changing with time, thereby obtaining a trajectory, and the trajectory is taken as the working device trajectory of the earth-moving machine. Each point on the working device trajectory is the pose space coordinates of the bucket 10 tooth tip at each time point.

[0158] Finally, based on the electronic map of the working environment and the working device trajectory of the earth-moving machine, the worked earth volume is determined. In this embodiment, the worked earth volume can be obtained by calculating the intersection based on the electronic map of the working environment and the working device trajectory of the earth-moving machine, and specifically can be obtained by the following steps:

[0159] Firstly, the to-be-worked range in the electronic map of the working environment is divided into a plurality of uniform cubes;

[0160] Secondly, the intersection state of each cube with the working device trajectory is determined respectively to obtain a plurality of cube states; in this embodiment, each cube state can be determined respectively according to the working device trajectory and the preset cube state judgment formula; in this embodiment, the pose space coordinates in the working device trajectory are substituted into the cube state judgment formula to obtain each cube state. Wherein, the cube state judgment formula is:

[0161]

[0162] In the formula: M j is a cube state, M j =1 indicates that it has been excavated, M j =0 indicates that it has not been excavated; V i is the space range defined by the cube; [x, y, z, ζ] is the pose space coordinates in the working device trajectory, and j is the number of the cube.

[0163] Thirdly, according to each cube state and the volume of each cube, the worked earth volume is obtained. In this embodiment, each cube state and the volume of each cube can be substituted into the following formula:

[0164] V=V0×ΣM j ;

[0165] In the formula: V0 is the volume of each cube, M j is each cube state, and V is the worked earth volume.

[0166] In the implementation process, the terrain information of the working range of the earth-moving machine is acquired, and an operation environment electronic map is formed according to the terrain information; the positioning signal of the earth-moving machine and the motion information of the actuator are acquired in real time; then the position of the earth-moving machine is determined in the operation environment electronic map according to the positioning signal of the earth-moving machine, and the working device track of the earth-moving machine is obtained according to the motion information of the actuator of the earth-moving machine; finally, the worked earth volume is obtained according to the state of each cube and the volume of each cube. The worked earth volume of the earth-moving machine is counted in the spatial dimension, and the earth volume is determined by the intersection based on the electronic map and the motion track, so that the earth operation volume of a single earth-moving machine can be obtained more conveniently, so as to evaluate the efficiency of a single earth-moving machine.

[0167] After obtaining the worked earth volume, the earth operation efficiency in the spatial dimension can be further counted, specifically: the earth operation efficiency is calculated according to the worked earth volume and the total operation time of earth excavation in the first operation information counting result; in this embodiment, the worked earth volume and the total operation time of earth excavation in the first operation information counting result are substituted into the following formula:

[0168]

[0169] In the formula, η is the earth operation efficiency, V is the worked earth volume, and T21 is the total operation time of earth excavation.

[0170] Correspondingly, after obtaining the worked earth volume, the single earth excavation operation volume in the spatial dimension can be further counted, specifically: the single earth excavation operation volume is calculated according to the worked earth volume and the earth excavation cycle number in the working condition information counting result. In this embodiment, the worked earth volume and the earth excavation cycle number in the working condition information counting result are substituted into the following formula:

[0171]

[0172] In the formula, V is the worked earth volume, is the single earth excavation operation volume, and N1 is the earth excavation cycle number; wherein, there are five working condition stages for earth excavation, and one cycle is calculated for one time of completion of the five working condition stages.

[0173] In the above implementation process, by statistically analyzing the trajectory of the working device, the working condition information, and the first operational information, the operational information of the earthmoving machinery is statistically analyzed in the spatial dimension to obtain the amount of earthwork already completed, earthmoving efficiency, and the amount of earthwork excavation per cycle, thereby achieving an evaluation of the operational efficiency of a single earthmoving machine. This enables the statistical analysis of operational information in the spatial dimension, providing a data foundation for host machine health evaluation and intelligent construction. Simultaneously, it allows for the acquisition of statistical information on earthmoving volume and the operational efficiency of a single machine, further facilitating project managers in making reasonable project decisions.

[0174] The completed earthwork volume, earthwork efficiency, and single excavation volume can be used as the second set of operational information statistics. After obtaining the first and second sets of operational information statistics, multi-level and multi-dimensional operational information statistics can be achieved. These statistics cover three aspects: working conditions, energy consumption, and earthwork volume, providing more comprehensive information on the host machine's operational process and supporting project managers in making reasonable project decisions. Furthermore, the first and second sets of operational information statistics can be uploaded to an IoT platform via an external controller for remote monitoring of the host machine's operational information.

[0175] Figure 1 This is a flowchart illustrating the earthmoving machinery condition identification method in this embodiment. It should be understood that, although... Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0176] In one embodiment, such as Figure 5 As shown, Figure 5 This schematically illustrates a structural block diagram of an earthmoving machinery condition identification device according to an embodiment of this application. The earthmoving machinery condition identification device includes an acquisition module 410, a first identification module 420, and a second identification module 430, wherein:

[0177] The acquisition module 410 is used to acquire basic data of earthmoving machinery; wherein, the data type of the basic data includes at least one of the following: main pump pressure signal, data acquired by machine vision, pilot control signal of operating handle, and displacement signal of actuator;

[0178] The first identification module 420 is configured to identify the working condition stage of the earth-moving machine according to the basic data by using a preset working condition stage identification model, to obtain a working condition stage identification result.

[0179] The second identification module 430 is configured to identify the working condition type of the earth-moving machine according to the working condition stage identification result by using a preset working condition type identification model, to obtain a working condition type identification result.

[0180] The earth-moving machine working condition identification device includes a processor and a memory, and the above-mentioned acquisition module 410, the first identification module 420 and the second identification module 430 are all stored in the memory as program units, and the corresponding functions are realized by the processor executing the above-mentioned program modules stored in the memory.

[0181] The processor includes a core, and the core calls the corresponding program units from the memory. The core can be set to one or more, and the multi-level working condition identification is realized by adjusting the core parameters.

[0182] The memory can include a non-permanent memory in a computer readable medium, a random access memory (RAM) and / or a non-volatile memory such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory includes at least one memory chip.

[0183] The embodiment of the present application provides a storage medium, which has a program stored thereon, and the program is executed by a processor to realize the above-mentioned earth-moving machine working condition identification method.

[0184] In one embodiment, a computer device is provided, which can be a terminal, and an internal structure diagram of the computer device can be as shown in Figure 6As shown in the figure, the computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05, and a memory (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A06. The network interface A02 is used for communication with external terminals via a network connection. When the computer program is executed by the processor A01, it implements a method for identifying the working conditions of earthmoving machinery. The display screen A04 can be a liquid crystal display (LCD) or an e-ink display. The input device A05 can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0185] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0186] In one embodiment, the earthmoving machinery condition identification device provided in this application can be implemented as a computer program, which can be implemented in the form of, for example... Figure 6 The device runs on the computer shown. The computer's memory can store the various program modules that make up the earthmoving machinery condition identification device, for example, Figure 5 The acquisition module 410, the first identification module 420, and the second identification module 430 are shown. The computer program composed of these program modules causes the processor to execute the steps in the earthmoving machinery working condition identification method of the various embodiments of this application described in this specification.

[0187] Figure 6 The computer device shown can be used as follows Figure 5 The acquisition module 410 in the earthmoving machinery condition identification device shown executes step 210. The computer device can execute step 220 through the first identification module 420 and step 230 through the second identification module 430.

[0188] This application provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps:

[0189] Obtaining basic data of the earth-moving machine;

[0190] According to the basic data, a preset working condition stage recognition model is used to recognize the working condition stage of the earth-moving machine, to obtain a working condition stage recognition result;

[0191] According to the working condition stage recognition result, a preset working condition type recognition model is used to recognize the working condition type of the earth-moving machine, to obtain a working condition type recognition result;

[0192] The data types of the basic data at least include one of a main pump pressure signal, data collected by machine vision, an operation handle pilot control signal, and an actuator displacement signal.

[0193] In one embodiment, the construction process of the preset working condition stage recognition model includes:

[0194] Obtaining first sample data, which includes main pump pressure data of each working condition stage under all working condition types, and working condition stage labels corresponding to the main pump pressure data;

[0195] The main pump pressure data of each working condition stage is respectively input into a first neural network, to obtain a predicted working condition stage;

[0196] The parameters of the first neural network are adjusted according to the predicted working condition stage and the working condition stage labels corresponding to the main pump pressure data in the first sample data, to obtain a working condition stage recognition model.

[0197] In one embodiment, the construction process of the preset working condition type recognition model includes:

[0198] Obtaining second sample data, which includes feature data of each working condition stage under all working condition types, and working condition type labels corresponding to the feature data;

[0199] The feature data of each working condition stage is respectively input into a second neural network, to obtain a predicted working condition type;

[0200] The parameters of the second neural network are adjusted according to the predicted working condition type and the working condition type labels corresponding to the feature data in the second sample data, to obtain a working condition type recognition model.

[0201] In one embodiment, the basic data is a main pump pressure signal, and the main pump pressure signal includes a plurality of main pump pressure data within a first preset time range from a current time;

[0202] According to the basic data, a preset working condition stage recognition model is used to recognize the working condition stage of the earth-moving machine, to obtain a working condition stage recognition result, including:

[0203] A1: feature extraction is performed on the main pump pressure signal to obtain a feature vector;

[0204] A2: normalization processing is performed on the feature vector to obtain a normalized feature vector;

[0205] A3: the normalized feature vector is input into a preset working condition stage recognition model for working condition stage recognition to obtain a working condition stage recognition result;

[0206] A4: it is judged whether the working condition stage recognition process is ended;

[0207] A5: in the case where the working condition stage recognition process is ended, the working condition stage recognition result is output;

[0208] A6: in the case where the working condition stage recognition process is not ended, the main pump pressure data farthest from the current time in the main pump pressure signal is eliminated, and the main pump pressure data at the next time is acquired to update the main pump pressure signal to obtain a new main pump pressure signal, and A1 is executed again.

[0209] In one embodiment, according to the working condition stage recognition result, a preset working condition type recognition model is used to recognize the working condition type of the earth-moving machine to obtain a working condition type recognition result, which comprises:

[0210] B1: a plurality of working condition stage recognition results within a second preset time range from the current time are acquired, and the plurality of working condition stage recognition results are combined to form a working condition stage vector;

[0211] B2: the working condition stage vector is input into a preset working condition type recognition model for working condition type recognition to obtain a working condition type recognition result;

[0212] B3: it is judged whether the working condition type recognition process is ended;

[0213] B4: in the case where the working condition type recognition process is ended, the working condition type recognition result is output;

[0214] B5: in the case where the working condition type recognition process is not ended, the working condition stage recognition result farthest from the current time in the plurality of working condition stage recognition results is eliminated, and the working condition stage recognition result at the next time is acquired to update the working condition stage vector to obtain a new working condition stage vector, and B2 is executed again.

[0215] In one embodiment, after obtaining the working condition type recognition result, it further comprises:

[0216] According to the working condition stage recognition result and the working condition type recognition result, statistics are performed according to a preset working condition information statistical condition to obtain a working condition information statistical result.

[0217] acquire fuel consumption rate information;

[0218] According to the working condition stage identification result, the working condition type identification result, the working condition information statistical result and the fuel consumption rate information, the working information of the earth-moving machine is statistically obtained in a time dimension to obtain a first working information statistical result.

[0219] In one embodiment, further comprising:

[0220] acquire terrain information of a range to be worked by the earth-moving machine, and form an electronic map of a working environment according to the terrain information;

[0221] acquire a positioning signal and execution mechanism movement information of the earth-moving machine in real time;

[0222] determine the position of the earth-moving machine in the electronic map of the working environment according to the positioning signal of the earth-moving machine, and obtain a working device track of the earth-moving machine according to the execution mechanism movement information of the earth-moving machine;

[0223] determine the worked earth volume based on the electronic map of the working environment and the working device track of the earth-moving machine;

[0224] The working device track is a track formed by the execution mechanism movement information of the earth-moving machine changing over time.

[0225] In one embodiment, the determination of the worked earth volume based on the electronic map of the working environment and the working device track of the earth-moving machine comprises:

[0226] divide the range to be worked in the electronic map of the working environment into a plurality of uniform cubes;

[0227] determine the intersection state of each cube with the working device track respectively to obtain a plurality of cube states;

[0228] obtain the worked earth volume according to the cube states and the volume of each cube.

[0229] In one embodiment, further comprising at least one of the following steps:

[0230] calculate the earth working efficiency according to the worked earth volume and the total working time of earth excavation, and the first working information statistical result at least includes the total working time of earth excavation;

[0231] calculate the single earth excavation working volume according to the worked earth volume and the earth excavation cycle number, and the working condition information statistical result at least includes the earth excavation cycle number.

[0232] Those skilled in the art will appreciate that embodiments of the application can be readily used as software, hardware, or a combination of software and hardware. In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0233] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing the one or more functions specified in the flowchart block or blocks.

[0234] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing the one or more functions specified in the flowchart block or blocks.

[0235] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing the one or more functions specified in the flowchart block or blocks.

[0236] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0237] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) about which the processor can execute instructions. The memory can also include non-volatile memory, such as read only memory (ROM) or flash RAM, about which permanent information can be stored; such information can not change much and / or can only change slowly.

[0238] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0239] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0240] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A method for identifying the working conditions of earthmoving machinery, characterized in that, include: Obtain basic data for earthmoving machinery; Based on the aforementioned basic data, a pre-set working condition stage identification model is used to identify the working condition stages of the earthmoving machinery, thereby obtaining the working condition stage identification results. Based on the working condition stage identification results, a preset working condition type identification model is used to identify the working condition type of the earthmoving machinery, and the working condition type identification results are obtained. The output of the working condition stage identification model serves as the input of the working condition type identification model. The data type of the basic data includes at least one of the following: main pump pressure signal, data acquired using machine vision, pilot control signal of operating handle, and displacement signal of actuator.

2. The earthmoving machinery operating condition identification method according to claim 1, characterized in that, The construction process of the working condition stage identification model includes: Obtain the first sample data, which includes the main pump pressure data of each operating condition stage under each operating condition type, and the operating condition stage label corresponding to the main pump pressure data. The main pump pressure data for each operating condition stage are input into the first neural network to obtain the predicted operating condition stage. The parameters of the first neural network are adjusted according to the predicted operating condition stage and the operating condition stage label corresponding to the main pump pressure data in the first sample data to obtain the operating condition stage identification model.

3. The earthmoving machinery operating condition identification method according to claim 1, characterized in that, The construction process of the working condition type identification model includes: Obtain second sample data, which includes feature data of each working condition stage under each working condition type, and working condition type labels corresponding to the feature data; The feature data of each working condition stage are input into the second neural network to obtain the predicted working condition type; The parameters of the second neural network are adjusted according to the predicted working condition type and the working condition type label corresponding to the feature data in the second sample data to obtain the working condition type recognition model.

4. The earthmoving machinery operating condition identification method according to claim 1, characterized in that, The basic data is the main pump pressure signal, which includes multiple main pump pressure data within a first preset time range from the current time. The step involves identifying the working stages of the earthmoving machinery using a pre-set working stage identification model based on the aforementioned basic data, thereby obtaining the working stage identification results, including: A1: Perform feature extraction on the main pump pressure signal to obtain a feature vector; A2: Normalize the feature vector to obtain the normalized feature vector; A3: Input the normalized feature vector into the preset working condition stage identification model to identify the working condition stage and obtain the working condition stage identification result. A4: Determine whether the working condition stage identification process has ended; A5: Output the working condition stage identification result after the working condition stage identification process is completed; A6: If the identification process of the working condition stage has not ended, remove the main pump pressure data that is furthest from the current time in the main pump pressure signal, and obtain the main pump pressure data of the next time to update the main pump pressure signal, so as to obtain a new main pump pressure signal, and return to execute A1.

5. The earthmoving machinery operating condition identification method according to claim 1, characterized in that, The step of identifying the working condition type of the earthmoving machinery using a pre-set working condition type identification model based on the working condition stage identification result, and obtaining the working condition type identification result, includes: B1: Obtain multiple operating condition stage identification results within a second preset time range from the current time, and combine the multiple operating condition stage identification results into an operating condition stage vector; B2: Input the working condition stage vector into the preset working condition type recognition model to identify the working condition type and obtain the working condition type recognition result; B3: Determine whether the working condition type identification process has ended; B4: Output the working condition type identification result after the working condition type identification process is completed; B5: If the working condition type identification process is not completed, remove the working condition stage identification result that is furthest from the current time among the multiple working condition stage identification results, obtain the working condition stage identification result of the next time moment to update the working condition stage vector, obtain a new working condition stage vector, and return to execute B2.

6. The earthmoving machinery operating condition identification method according to claim 1, characterized in that, After obtaining the operating condition type identification result, the following is also included: Based on the identification results of the working condition stage and the identification results of the working condition type, statistics are performed according to the preset working condition information statistical conditions to obtain the working condition information statistical results. Obtain fuel consumption rate information; Based on the working condition stage identification results, working condition type identification results, working condition information statistics results, and fuel consumption rate information, the operation information of the earthmoving machinery is statistically analyzed in the time dimension to obtain the first operation information statistics result.

7. The earthmoving machinery operating condition identification method according to claim 6, characterized in that, Also includes: Obtain terrain information of the area where earthmoving machinery will operate, and generate an electronic map of the operating environment based on the terrain information; Real-time acquisition of positioning signals and actuator motion information of earthmoving machinery; Based on the positioning signal of the earthmoving machinery, the location of the earthmoving machinery is determined in the electronic map of the working environment, and the trajectory of the working device of the earthmoving machinery is obtained based on the motion information of the actuator of the earthmoving machinery. Based on the electronic map of the work environment and the working device trajectory of the earthmoving machinery, the amount of earthwork already done is determined. The trajectory of the working device is the trajectory formed by the change of motion information of the actuator of the earthmoving machinery over time.

8. The earthmoving machinery operating condition identification method according to claim 7, characterized in that, The determination of the amount of earthwork already completed based on the electronic map of the working environment and the trajectory of the earthmoving machinery includes: The work area in the electronic map of the work environment is divided into multiple uniform cubes; The intersection state between each cube and the trajectory of the working device is determined to obtain multiple cube states; Based on the state and volume of each cube, the amount of earthwork already completed is obtained.

9. The earthmoving machinery operating condition identification method according to claim 7, characterized in that, It also includes at least one of the following steps: Based on the amount of earthwork already completed and the total operation time of earthwork excavation, the earthwork operation efficiency is calculated, and the first operation information statistics result includes at least the total operation time of earthwork excavation. Based on the amount of earthwork already done and the number of earthwork excavation cycles, the amount of earthwork excavation work per cycle is calculated, and the statistical results of the working condition information include at least the number of earthwork excavation cycles.

10. A device for identifying the working condition of earthmoving machinery, characterized in that, include: The acquisition module is used to acquire basic data of earthmoving machinery; wherein, the data type of the basic data includes at least one of the following: main pump pressure signal, data acquired by machine vision, pilot control signal of operating handle, and displacement signal of actuator; The first identification module is used to identify the working stage of the earthmoving machinery based on the basic data and using a preset working stage identification model to obtain the working stage identification result. The second identification module is used to identify the working condition type of the earthmoving machinery based on the working condition stage identification result and a preset working condition type identification model to obtain the working condition type identification result; wherein, the output of the working condition stage identification model is used as the input of the working condition type identification model.

11. A processor, characterized in that, It is configured to perform the earthmoving machinery condition identification method according to any one of claims 1 to 9.

12. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the earthmoving machinery condition identification method according to any one of claims 1 to 9.

Images