KR102483801B1 - Method, apparatus and system for controlling wheel loader - Google Patents

Abstract

In a control method of a wheel loader, signals indicating a working state are received from sensors mounted on the wheel loader. Among the received signals, a signal required for determining each of a plurality of individual loads classified according to loads required in a series of tasks performed by the wheel loader is selected. A pre-learned prediction algorithm is performed on the selected signals to calculate output values indicating whether or not the plurality of individual load states are present. The current work load state is determined by analyzing the calculated output values.

Description

Wheel loader control method, control device and control system {METHOD, APPARATUS AND SYSTEM FOR CONTROLLING WHEEL LOADER}

The present invention relates to a control method, control device, and control system of a wheel loader, and more particularly, to a control method of a wheel loader for automatically controlling the wheel loader by determining a working state of the wheel loader, and a control for performing the same It relates to devices and control systems.

Wheel loaders are widely used at construction sites to excavate and transport soil, sand, etc., and to load them onto cargo vehicles such as dump trucks.

The work load is changed according to the working state of the wheel loader, and the engine or transmission of the wheel loader is automatically controlled by detecting the work load, thereby reducing fuel consumption and preventing work efficiency from being lowered. Therefore, a technique for accurately detecting the current work state and work load state in real time and automatically controlling the wheel loader based thereon is required.

An object of the present invention is to provide a control method of a wheel loader capable of reducing fuel consumption and improving work efficiency when the wheel loader works.

Another object of the present invention is to provide a control device for performing the above-described control method of a wheel loader.

Another object of the present invention is to provide a control system for performing the control method of the wheel loader described above.

In order to achieve the above-described object of the present invention, in the method for controlling a wheel loader according to exemplary embodiments of the present invention, signals indicating a working state are received from sensors mounted on the wheel loader. Among the received signals, a signal required for determining each of a plurality of individual loads classified according to loads required in a series of tasks performed by the wheel loader is selected. A pre-learned prediction algorithm is performed on the selected signals to calculate output values indicating whether or not the plurality of individual load states are present. The current work load state is determined by analyzing the calculated output values.

In example embodiments, performing the pre-learned prediction algorithm may include performing a neural network algorithm.

In example embodiments, the calculating of the output values indicating whether or not the plurality of individual load states is performed may include an output value indicating whether a low load state is present, an output value indicating whether a heavy load state is present, an output value indicating whether a high load state is present, and an acceleration/acceleration / A step of calculating an output value indicating whether or not a slope load state may be included.

In exemplary embodiments, at least one of a boom cylinder pressure signal, an FNR signal, a main pressure signal of a hydraulic pump, a vehicle speed signal, a boom position signal, and a torque converter speed ratio signal is a low load state and a high load state of the wheel loader At least one of the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal may be used to determine whether the wheel loader is in a heavy load state.

In example embodiments, at least one of a torque converter speed ratio signal and an accelerator pedal position signal may be used to determine whether the wheel loader is in an acceleration/slope load state.

In exemplary embodiments, forward travel, reverse travel, dump work, and reverse boom down operation of the V-shaped driving of the wheel loader are determined as a low load state, and an excavation work is determined as a heavy load state, , the forward traveling boom-up task may be determined as a high load state.

In example embodiments, the method may further include outputting a control signal for controlling an engine or transmission of the wheel loader according to a current work load state of the wheel loader.

In order to achieve the other object of the present invention described above, a control device for a wheel loader according to exemplary embodiments of the present invention includes a signal receiver for receiving signals indicating work state information from sensors mounted on the wheel loader, the wheel A signal selection unit that classifies a series of tasks performed by the loader into a plurality of individual load states according to the load state required and selects and inputs a signal necessary to determine each of the individual load states among the received signals , an individual load determination unit which calculates an output value representing each of the individual load conditions by performing a pre-learned prediction algorithm on the selected signals, and a load status determination unit which analyzes the output values and determines a current workload status do.

In example embodiments, the individual load determination unit may include individual determination circuit units that perform a neural network algorithm.

In exemplary embodiments, the individual load determination unit may include a low load determination circuit unit calculating an output value indicating whether a low load state exists, a heavy load determination circuit unit calculating an output value indicating whether a heavy load state exists, and an output value indicating whether a high load state exists. It may include a high load determination circuit unit that calculates, and an acceleration/slope load determination circuit unit that calculates an output value indicating whether or not an acceleration/slope load condition exists.

In exemplary embodiments, the low load determination circuit unit and the high load determination circuit unit receive at least one of a boom cylinder pressure signal, an FNR signal, a main pressure signal of a hydraulic pump, a vehicle speed signal, a boom position signal, and a torque converter speed ratio signal. to determine whether the wheel loader is in a low load state or a high load state, respectively, and the heavy load determination circuit unit determines at least one of the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal. One can be used to determine whether the wheel loader is in a heavy load state.

In example embodiments, the acceleration/slope load determination circuit unit may determine whether the wheel loader is in an acceleration/slope load state by using at least one of a torque converter speed ratio signal and an accelerator pedal position signal.

In exemplary embodiments, the load state determination unit determines that a forward driving operation, a backward traveling operation, a dump operation, and a reverse traveling boom-down operation in the V-shaped operation of the wheel loader are in a low load state, and the excavation operation is performed. It may be determined as a heavy load state, and a forward driving boom-up operation may be determined as a high load state.

In exemplary embodiments, the control device of the wheel loader may further include a control signal output unit for outputting a control signal for controlling an engine or transmission of the wheel loader according to a current work load state of the wheel loader. .

In order to achieve another object of the present invention described above, a control system for a wheel loader according to exemplary embodiments of the present invention includes an engine, a work device and a travel device driven by the engine, the engine, the work device, and A plurality of sensors mounted on the traveling device to detect signals representing work state information of the wheel loader, and a load state required by a series of tasks performed by the wheel loader among signals received from the sensors Control for determining the current workload state by selecting signals capable of determining each of a plurality of individual load states, and determining whether or not the individual load states are present by performing a pre-learned prediction algorithm for the selected signals. include the device

In example embodiments, the control device selects a signal necessary for determining the individual load state from among the received signals, calculates an output value indicating each of the individual load state, and analyzes the output values to determine the individual load state. Work load status can be judged.

In example embodiments, the control device may calculate the output value by performing a neural network algorithm on the selected signals.

In exemplary embodiments, the control device calculates an output value indicating whether a low load state exists, calculates an output value indicating whether a heavy load state exists, calculates an output value indicating whether a high load state exists, and calculates an acceleration/slope load state. It is possible to calculate an output value indicating whether or not.

In exemplary embodiments, the control device determines that forward travel, reverse travel, dump work, and reverse boom down operation in the V-type operation of the wheel loader are in a low load state, and the excavation operation is performed in a heavy load state. It is determined as a low state, and the forward traveling boom-up operation may be determined as a high load state.

In example embodiments, the control device may output a control signal for controlling the engine or transmission of the wheel loader according to a current work load state of the wheel loader.

According to exemplary embodiments, signals representing the working state of the wheel loader are received and an individual load state (low load, medium load, high load, acceleration/slope load) is best represented among specific signals among the received signals. It is possible to selectively select possible signals and determine the current work load state or current work state of the wheel loader using a pre-learned prediction algorithm such as a neural network algorithm.

Accordingly, it is possible to reduce the time and burden required for calculating the work load state of the wheel loader and improve the accuracy of the determination. In addition, it is possible to improve work performance and fuel efficiency by efficiently controlling the engine and transmission according to the finally determined work load state.

However, the effects of the present invention are not limited to the above-mentioned effects, and may be variously extended without departing from the spirit and scope of the present invention.

1 is a side view illustrating a wheel loader according to exemplary embodiments.
Figure 2 is a block diagram showing a control system of the wheel loader of Figure 1;
Fig. 3 is a block diagram showing a control device of a wheel loader according to exemplary embodiments.
FIG. 4 is a block diagram illustrating a signal selection unit, an individual load determination unit, and a load state determination unit of the control device of FIG. 3 .
FIG. 5 is a diagram illustrating an individual neural network circuit of an individual load determination unit of FIG. 4 .
FIG. 6 is a diagram showing signal transmission equations in each layer of the individual neural network circuit of FIG. 5 .
7 is a flowchart illustrating a method of controlling a wheel loader according to exemplary embodiments.
8 is a diagram illustrating V-shaped operation of a wheel loader according to exemplary embodiments.
FIG. 9 is graphs showing output values representing individual load states in each operation of the V-shaped driving of FIG. 8 .
10 is a graph showing a final workload state obtained by analyzing the output values of FIG. 9 .

For the embodiments of the present invention disclosed in the text, specific structural or functional descriptions are only exemplified for the purpose of explaining the embodiments of the present invention, and the embodiments of the present invention may be implemented in various forms and the text It should not be construed as being limited to the embodiments described above.

Since the present invention can have various changes and various forms, specific embodiments will be illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to a specific form disclosed, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms may be used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle. Other expressions describing the relationship between elements, such as "between" and "directly between" or "adjacent to" and "directly adjacent to", etc., should be interpreted similarly.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as "comprise" or "having" are intended to indicate that the described feature, number, step, operation, component, part, or combination thereof exists, but that one or more other features or numbers are present. However, it should be understood that it does not preclude the presence or addition of steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in this application, they are not interpreted in an ideal or excessively formal meaning. .

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described in more detail. The same reference numerals are used for the same components in the drawings, and redundant descriptions of the same components are omitted.

1 is a side view illustrating a wheel loader according to exemplary embodiments. Figure 2 is a block diagram showing a control system of the wheel loader of Figure 1;

Referring to FIGS. 1 and 2 , the wheel loader 10 may include a front body 12 and a rear body 14 rotatably connected to each other. The front body 12 may include a working device and a front wheel 160 . The rear body 14 may include a cab 40 , an engine room 50 and rear wheels 162 .

The work device may include a boom 20 and a bucket 30 . The boom 20 may be freely rotatably attached to the front vehicle body 12 and the bucket 30 may be freely rotatably attached to one end of the boom 20 . The boom 20 is connected to the front vehicle body 12 by a pair of boom cylinders 22, and the boom 20 can rotate vertically by driving the boom cylinder 22. The tilt arm 34 is freely rotatably attached almost at the center of the arm 20, and one end of the tilt arm 34 and the front body 12 are connected by a pair of bucket cylinders 32, The bucket 30 connected to the other end of the tilt arm 34 by a tilt rod can be rotated (dump or crowd) in an up and down direction by driving the bucket cylinder 32 .

In addition, the front body 12 and the rear body 14 are rotatably connected to each other by a center pin 16, and the front body 12 is rotatably connected to the rear body 14 by extension and contraction of a steering cylinder (not shown). It can be refracted left and right for .

A traveling device for driving the wheel loader 10 may be mounted on the rear body 14 . The engine 100 may be disposed in the engine room 50 and supply power output to the traveling device. The traveling device may include a torque converter 120, a transmission 130, a propeller shaft 150, and axles 152 and 154. The power output of the engine 100 is transmitted to the front wheel 160 and the rear wheel 162 through the torque converter 120, the transmission 130, the propeller shaft 150 and the axles 152 and 154, and the wheel loader ( 10) will run.

Specifically, power output of the engine 100 may be transmitted to the transmission 130 through the torque converter 120 . An input shaft of the torque converter 120 may be connected to an output shaft of the engine 100 and an output shaft of the torque converter 120 may be connected to the transmission 130 . The torque converter 120 may be a fluid clutch device having an impeller, turbine and stator. The transmission 130 may include hydraulic clutches for shifting speed stages between first and fourth speeds, and rotation of the output shaft of the torque converter 120 may be shifted by the transmission 130 . The shifted rotation is transmitted to the front wheel 160 and the rear wheel 162 through the propeller shaft 150 and the axles 152 and 154 so that the wheel loader can travel.

The torque converter 120 may have a function of increasing output torque with respect to input torque, that is, a function of making a torque ratio equal to or greater than 1. The torque ratio decreases as the torque converter speed ratio e (= Nt/Ni), which is the ratio between the number of revolutions Ni of the input shaft and the number of revolutions Nt of the output shaft of the torque converter 120, increases. For example, when the driving load increases while driving with the engine speed constant, the rotation speed of the output shaft of the torque converter 120, that is, the vehicle speed decreases, and the torque converter speed ratio decreases. At this time, since the torque ratio is increased, it is possible to travel with a greater travel driving force.

The transmission 130 may include a forward hydraulic clutch, a reverse hydraulic clutch, and first to fourth hydraulic clutches. Each of the hydraulic clutches may be engaged or released by hydraulic oil (clutch pressure) supplied through a transmission control unit (TCU) 140. That is, when the clutch pressure supplied to the hydraulic clutch increases, the hydraulic clutch can be engaged and released when the clutch pressure decreases.

When the driving load decreases and the torque converter speed ratio (e) increases and exceeds the preset value (eu), the speed stage is shifted up by one stage. Conversely, when the driving load increases and the torque converter speed ratio (e) becomes less than the preset value (ed), the speed stage is shifted down by one stage.

The transmission 130 may have a manual mode or a plurality of automatic shift modes. The shift modes may be changed by operating a mode conversion switch (not shown). For example, the transmission 130 may include a manual mode, a 1-4 auto mode, and a 1-3 auto mode. When set to the manual mode, the speed stage selected by the shift selection lever may be applied. When the 1-4 auto mode or the 1-3 auto mode is set, the speed range may be automatically shifted between the speed ranges lower than the speed range selected by the shift selection lever.

A variable displacement hydraulic pump 200 for supplying hydraulic oil to the boom cylinder 22 and the bucket cylinder 32 of the working device may be mounted on the rear body 14 . Variable displacement hydraulic pump 200 can be driven using a portion of the power output from engine 100 . For example, the output of the engine 100, the hydraulic pump 200 for the work device through a power transmission device (PTO) such as a gear train 110 installed between the engine 100 and the torque converter 120 and a hydraulic pump (not shown) for steering may be driven.

A pump control device is connected to the variable displacement hydraulic pump 200, and a discharge flow rate of the variable displacement hydraulic pump 200 can be controlled by the pump control device. A main control valve MCV such as a boom control valve 210 and a bucket control valve 212 may be installed on the hydraulic circuit of the hydraulic pump 200 . The oil discharged from the hydraulic pump 200 is supplied to the boom cylinder 22 and the bucket cylinder 32 through the boom control valve 210 and the bucket control valve 212 installed in the hydraulic line 202 in front of the main control valve It can be. The main control valve (MCV) may supply hydraulic oil discharged from the hydraulic pump 200 to the boom cylinder 22 and the bucket cylinder 32 according to the pilot pressure input from the control lever. Accordingly, the boom 20 and the bucket 30 may be driven by hydraulic pressure of hydraulic oil discharged from the hydraulic pump 200 .

A driving control device may be provided in the cab 40 . The drive control device may include a travel pedal 142, a brake pedal 144, and operation levers for operating cylinders such as an FNR travel lever, a boom cylinder 22, and a bucket cylinder 32.

As described above, the wheel loader 10 is a driving system for driving the traveling device using the output of the engine 100 through a power transmission device (PTO) and work devices such as the boom 20 and the bucket 30 It may include a hydraulic system for driving.

In addition, the control device 300 of the wheel loader 10 may be mounted on the rear body 14 as a part of the vehicle control unit (VCU) or as a separate controller. The control device 300 may include an arithmetic processing device having a CPU that executes a program, a storage device such as a memory, and other peripheral circuits.

The control device 300 may receive signals from various sensors mounted on the wheel loader 10 . For example, the control device 300 includes an engine speed sensor 102 that detects engine speed, a travel pedal detection sensor 143 that detects an operation amount of a travel pedal 142, and an operation amount of a brake pedal 144. A brake pedal detection sensor 145 that detects, a speed stage of the transmission 130, an FNR lever position detection sensor 146 that detects the operation position of the FNR lever that selects forward (F), neutral (N), and reverse (R). ) can be connected to

In addition, the control device 300 includes a rotational speed detection sensor 122a for detecting the rotational speed (Ni) of the input shaft of the torque converter 120 and a rotational speed (Nt) of the output shaft of the torque converter 120. It may be connected to the rotational speed detection sensor 122b and the vehicle speed detection sensor 132 that detects the rotational speed of the output shaft of the transmission 130, that is, the vehicle speed v.

In addition, the control device 300 is installed in the hydraulic line 202 in front of the main control valve (MCV), the pressure sensor 204 for detecting the discharge pressure of the hydraulic pump 200, and the head side of the boom cylinder 22 It may be connected to a boom cylinder pressure sensor 222 that detects pressure. In addition, the control device 300 may be connected to the boom angle sensor 224 for detecting the rotational angle of the boom 20 and the bucket angle sensor 234 for detecting the rotational angle of the bucket 30 .

Signals detected by the sensors mounted on the wheel loader 10 may be input to the control device 300 as indicated by the dotted arrow in FIG. 2 . As will be described later, the control device 300 selects specific signals from among signals received from sensors mounted on the wheel loader 10 and performs a pre-learned prediction algorithm such as a neural network algorithm to determine whether each individual load condition is present. It is possible to determine the current work load state or the current work state of the wheel loader 10 by calculating and analyzing output values representing . Furthermore, the control unit 300 may be connected to an engine control unit (ECU), a transmission control unit (TCU) 140, the pump control unit, etc. to output a control signal, and considering the final load state or work state, the wheel The engine 100, the transmission 130, the hydraulic pump 200, and the like of the loader 10 may be selectively controlled.

Hereinafter, a control device of the wheel loader will be described.

Fig. 3 is a block diagram showing a control device of a wheel loader according to exemplary embodiments. FIG. 4 is a block diagram illustrating a signal selection unit, an individual load determination unit, and a load state determination unit of the control device of FIG. 3 . FIG. 5 is a diagram illustrating an individual neural network circuit of an individual load determination unit of FIG. 4 . FIG. 6 is a diagram showing signal transmission equations in each layer of the individual neural network circuit of FIG. 5 .

Referring to FIGS. 3 to 6 , the wheel loader control device 300 may include a work load determination unit 310 , a control signal output unit 320 and a storage unit 330 .

The work load determination unit 310 may determine a load state of a current work performed by the wheel loader 10 or a current work state received from sensors mounted on the wheel loader 10 . The control signal output unit 320 may select a type of control to be performed, such as, for example, engine output torque control, engine rpm control, transmission shift control, etc., according to the determined current work load or work state. . The storage unit 330 stores data necessary for calculations such as training for a predictive model performed by the workload determination unit 310 and performing a neural network algorithm, and a control map necessary for determining a control signal in the control signal output unit 320. etc. can be stored.

In example embodiments, the workload determination unit 310 may include a signal reception unit 312 , a signal selection unit 314 , an individual load determination unit 316 and a load condition determination unit 318 .

The signal receiver 312 may receive signals indicating a work state from sensors mounted on the wheel loader 10 . For example, the signal receiving unit 312 is a boom cylinder pressure signal from the pressure sensor 222 installed on the head side of the boom cylinder 22, the FNR signal from the FNR lever position detection sensor 146, and the discharge of the hydraulic pump 200 The main pressure signal of the hydraulic pump from the pressure sensor 204, the vehicle speed signal from the vehicle speed detection sensor 132, the boom position signal from the boom angle sensor 224, and the rotation speed detection sensors 122a and 122b of the torque converter 120 The ratio of the number of rotations (Ni) of the input shaft and the number of rotations (Nt) of the output shaft obtained from , that is, a torque converter speed ratio signal and an accelerator pedal position signal from the travel pedal detection sensor 145 may be received. Signals received by the signal receiving unit 312 are not limited thereto, and it will be understood that various signals that can be used to determine the work load state or work state of the wheel load can be received.

The signal receiver 312 may include a data pre-processor. The data pre-processing unit may filter the input sensor signals to remove noise and normalize them.

The signal selector 314 selects a signal capable of determining a plurality of individual load states, for example, load states divided into at least four, among the received signals, and selects the selected signal as an individual load corresponding thereto. It may be output to each of the individual decision circuit units NN_1, NN_2, NN_3, and NN_4 of the decision unit 316. The signal selector 314 may select signals necessary for determining at least first to fourth individual load states, respectively, classified according to loads required in a series of tasks performed by the wheel loader. For example, the individual load states may be classified into a low load state, a heavy load state, a high load state, and an acceleration/slope state according to loads required in a series of tasks performed by the wheel loader.

The signal selected from among the received signals is a specific load state required in a specific task performed by the wheel loader 10, that is, at least one of a low load state, a heavy load state, a high load state, and an acceleration/slope load state It can be an indicator that can effectively indicate the load condition of

The boom cylinder pressure signal may be an indicator that can directly indicate the current work load state of the wheel loader according to the weight of the load loaded in the bucket 30, the height of the boom 20, and the like. The boom cylinder pressure signal may be used to determine a driving operation state and a combined operation (traveling boom-up) state of the wheel loader.

The FNR signal may be an indicator for distinguishing between work transitions, such as start of reverse work after completion of excavation work or conversion of forward or backward during driving work. The FNR signal may be used to determine a driving work state and a combined work (running boom-up) state of the wheel loader.

Unlike the boom cylinder pressure, the main pressure signal of the hydraulic pump, that is, the MCV input pressure, is maintained in a basic pressure state unless the driver manipulates the boom/bucket, so at least one of the excavation work state or the boom 20 and the bucket 30 It can be an indicator representing any one operation. The main pressure signal of the hydraulic pump may be used to determine the wheel loader's running work state, combined work (travel boom-up) state, and excavation work state.

The vehicle speed signal may be an index indicating the traveling speed of the wheel loader. The vehicle speed signal may be used to determine the wheel loader driving work state, complex work (driving boom-up) state, and excavation work state.

The boom position signal may be an indicator for distinguishing between operations of the boom position during driving and excavation work of the wheel loader and boom positions during vehicle dump work, which are different from each other. The boom position signal may be used to determine a driving work state, a combined work (travel boom up) state, and an excavation work state of the wheel loader.

The torque converter speed ratio signal is an index indicating a driving load of the vehicle and may indicate an excavation work state and a slope driving work state. The torque converter speed ratio signal may be used to determine a driving work state, a complex work (travel boom-up) state, an excavation work state, and an acceleration work state of the wheel loader.

The accelerator pedal position signal may be an indicator indicating the driver's intention to accelerate. The accelerator pedal position signal may be used to determine an accelerating operation state.

The individual load determination unit 316 may include a plurality of individual determination circuit units. Specifically, the individual load determination unit 316 may include first to fourth individual determination circuit units. The first to fourth individual determination circuit units may calculate output values representing the first to fourth individual load states, respectively, by performing a pre-learned prediction algorithm on the selected signals.

In exemplary embodiments, the first individual determination circuit unit may include a low load neural network determination unit NN_1 that calculates an output value indicating a low load state by performing a neural network algorithm on input signals. can The low load neural network determination unit NN_1 receives the boom cylinder pressure signal, the FNR signal, the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal from the signal selection unit 314 can receive The low load neural network determination unit NN_1 may calculate a first output value indicating whether the current operation of the wheel loader is in a low load state by performing a neural network algorithm on the received signals.

The second individual determination circuit unit may include a heavy load neural network determination unit NN_2 that calculates an output value indicating a heavy load state by performing a neural network algorithm on input signals. The heavy load neural network determination unit NN_2 may receive the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal from the signal selection unit 314 . The heavy load neural network determining unit NN_2 may calculate a second output value indicating whether the current operation of the wheel loader is in a heavy load state by performing a neural network algorithm on the received signals.

The third individual determination circuit unit may include a high load neural network determination unit NN_3 that calculates an output value indicating a high load state by performing a neural network algorithm on input signals. The high load neural network determination unit NN_3 receives the boom cylinder pressure signal, the FNR signal, the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal from the signal selection unit 314 can do. The high load neural network determination unit NN_3 may perform a neural network algorithm on the received signals to calculate a third output value indicating whether the current wheel loader operation is in a high load state.

The fourth individual determination circuit unit may include an acceleration/slope load neural network determination unit NN_4 that calculates an output value indicating an acceleration/slope load state by performing a neural network algorithm on input signals. The acceleration/slope load neural network determiner NN_4 may receive the torque converter speed ratio signal and the accelerator pedal position signal from the signal selector 314 . The acceleration/slope load neural network determining unit NN_4 may perform a neural network algorithm on the received signals to calculate a fourth output value indicating whether the wheel loader is currently in an acceleration/slope load state.

In exemplary embodiments, the low load neural network determiner NN_1, the heavy load neural network determiner NN_2, the high load neural network determiner NN_3, and the acceleration/slope load neural network determiner NN_4 are pre-learned neural network algorithms. Each may include a neural network circuit that calculates an output value representing an individual load state by performing

As shown in FIGS. 5 and 6 , the neural network circuit may have a multi-layer perceptron structure having a multi-layer input layer, a hidden layer, and an output layer. Neurons are arranged in each layer, and neurons in each layer can be connected by connection weights. That is, input data may be input to neurons of the input layer and transmitted to the output layer through the hidden layer.

Learning in the neural network algorithm may be a process of adjusting weights between nodes so that an error between an output value derived by the neural network algorithm and an expected value according to a specific input (actually measured data) is minimized. For example, the neural network algorithm of the neural network circuit may be learned by a back propagation learning method. Therefore, the neural network circuits of the individual neural network decision units NN_1, NN_2, NN_3, and NN_4 adjust the weights connecting the input layer, the hidden layer, and the output layer using data previously collected for each individual neural network decision unit to obtain a prediction model As a result, a neural network algorithm can be established.

Therefore, the neural network circuit has a pre-learned neural network algorithm, and may calculate an output value indicating whether or not an individual load state is present by performing such a neural network algorithm on input signals.

The load state determination unit 318 may determine the current work load state or work state of the wheel loader 10 by analyzing the first to fourth output values calculated by the first to fourth individual determination circuit units. The load state determination unit 318 may calculate a final result after performing post-processing such as weight application on the output values input from the individual neural network determination units NN_1, NN_2, NN_3, and NN_4.

For example, the load state determination unit 318 may determine the current work load state of the wheel loader 10 by analyzing the output values. Accordingly, the load state determination unit 318 may determine that the current work load state of the wheel loader 10 is any one of a low load state, a heavy load state, a high load state, and an acceleration/slope load state.

In addition, the load state determining unit 318 may determine the current working state of the wheel loader 10 by additionally considering other signals received from sensors mounted on the wheel loader 10 . For example, the load state determination unit 318 may determine the current work load state and the current work state of the wheel loader 10 .

The control signal output unit 320 controls the engine 100, transmission 130, and hydraulic pump of the wheel loader 10 in consideration of the current work load state or work state of the wheel loader determined by the work load determination unit 310. A control signal for selectively controlling the 200 and the like can be output. For example, the control signal output unit 320 may output a control signal to control engine output torque, engine rpm, the number of shift stages, shift timing, and the like.

Accordingly, the control signal output unit 320 may maintain or improve work performance and improve fuel efficiency by controlling the engine 100 and the transmission 130 in consideration of the finally determined work load state or work state.

The storage unit 330 is connected to the work load determination unit 310 and is connected to the first storage unit 332 for storing data for determining the workload and the control signal output unit 320 and stores data for the control signal. It may include a second storage unit 334 to determine. The first storage unit 332 may store data necessary for calculations such as learning for a prediction model and performing a neural network algorithm. The second storage unit 334 may store an engine torque map, an engine rpm map, a transmission shift control map, and the like necessary for determining a control signal.

As described above, the wheel loader control device 300 determines individual load conditions (low load, medium load, high load, acceleration/slope load) among specific signals among signals received from sensors mounted on the wheel loader 10 ), and the current work load state or current work state of the wheel loader 10 may be determined using a pre-learned prediction algorithm such as a neural network algorithm.

Accordingly, it is possible to reduce the time and burden required for calculating the work load state of the wheel loader and improve the accuracy of the determination. In addition, it is possible to improve work performance and fuel efficiency by efficiently controlling the engine and transmission according to the finally determined work load state.

Hereinafter, a method of controlling the wheel loader using the control device of the wheel loader of FIG. 3 will be described.

7 is a flowchart illustrating a method of controlling a wheel loader according to exemplary embodiments.

Referring to Figures 3, 4 and 7, first, signals indicating the working state of the wheel loader can be received (S100).

The control device 300 of the wheel loader may receive signals indicating a working state from sensors mounted on the wheel loader. For example, the signal receiving unit 312 of the work load determining unit 310 receives a boom cylinder pressure signal, an FNR signal, a main pressure signal of a hydraulic pump, a vehicle speed signal, a boom position signal, a torque converter speed ratio signal, and an accelerator pedal position signal. etc. can be received. The input sensor signals may be normalized after filtering through preprocessing.

Then, among the received signals, a signal necessary for determining each of a plurality of individual load states may be selected (S110).

The signal selection unit 314 of the work load determination unit 310 selects signals capable of determining the first to fourth individual load states divided into at least four from among the received signals, and responds to the selected signals. may be output to each of the individual determination circuit units NN_1, NN_2, NN_3, and NN_4 of the individual load determination unit 316.

The first to fourth individual load states may respectively correspond to a low load state, a heavy load state, a high load state, and an acceleration/slope state according to loads required in a series of tasks performed by the wheel loader. The received signals effectively represent at least one load condition among a specific load condition required for a specific task performed by the wheel loader, that is, a low load condition, a heavy load condition, a high load condition, and an acceleration/slope load condition. It can be classified according to availability.

For example, among the received signals, the boom cylinder pressure signal, the FNR signal, the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal indicate a low load of the wheel loader These signals may be signals necessary for determining whether the signals are in a high load state or not, and may be input to the low load neural network determining unit NN_1 and the high load neural network determining unit NN_3 of the individual load determining unit 316 .

Among the received signals, the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal may be signals necessary for determining whether the wheel loader is in a heavy load state, and individual load It may be input to the heavy load neural network determining unit NN_2 of the determining unit 316 .

Among the received signals, the torque converter speed ratio signal and the accelerator pedal position signal may be signals necessary for determining whether or not the acceleration/slope load state is present, and the acceleration/slope load neural network determination unit of the individual load determination unit 316 It can be input as (NN_4).

Thereafter, it is possible to determine each of a plurality of individual load states by performing a pre-learned prediction algorithm for the selected signals (S120).

The low load neural network determiner NN_1, heavy load neural network determiner NN_2, high load neural network determiner NN_3, and acceleration/slope load neural network determiner NN_4 of the individual load determiner 316 determine the selected signals. It is possible to calculate output values representing low load conditions, heavy load conditions, high load conditions, and acceleration/slope load conditions, respectively, by performing a pre-learned neural network algorithm.

Subsequently, the current work load state of the wheel loader may be determined by comprehensively analyzing the output values (S130).

The load state determination unit 318 analyzes the output values and determines that the current work load state of the wheel loader is any one of a low load state, a heavy load state, a high load state, and an acceleration/slope load state.

In addition, the load state determination unit 318 may additionally consider other signals received from sensors mounted on the wheel loader to determine the current work state as well as the current work load state of the wheel loader.

Subsequently, the engine, transmission, hydraulic pump, etc. of the wheel loader may be selectively controlled in consideration of the determined current work load state or work state of the wheel loader.

Hereinafter, a method of determining a workload state of V-shaped driving of the wheel loader using the wheel loader control method of FIG. 7 will be described.

8 is a diagram illustrating V-shaped operation of a wheel loader according to exemplary embodiments. FIG. 9 is graphs showing output values representing individual load states in each operation of the V-shaped driving of FIG. 8 . 10 is a graph showing a final workload state obtained by analyzing the output values of FIG. 9 . 9 and 10 together show graphs showing boom cylinder pressure values over time in the V-type operation for reference.

Referring to FIGS. 8 to 10 , the wheel loader 10 may perform a V-type operation of excavating a target material such as soil S and loading the dump truck T. For the V-type operation, the wheel loader 10 performs forward driving (a), excavation (b), reverse driving (c), forward driving boom-up operation (d), dump operation (e) and reverse driving. The boom-down operation (f) can be performed sequentially.

As shown in FIG. 9 , it is possible to determine individual load states corresponding to each task of the V-type driving. The low load neural network determination unit NN_1 may calculate an output value indicating whether or not the low load state is present for a series of tasks a to f. The heavy load neural network determination unit NN_2 may calculate an output value indicating whether or not the heavy load state is present for a series of tasks a to f. The high load neural network determination unit NN_3 may calculate an output value indicating whether a high load state is present for a series of tasks a to f. The acceleration/slope load neural network determination unit NN_4 may calculate an output value indicating whether or not the acceleration/slope load state is present for a series of tasks (a to f).

As shown in FIG. 10 , the current workload state may be determined by comprehensively analyzing the calculated output values. The load state determination unit 318 may determine that each of the series of tasks (a to f) has a low load state, a heavy load state, a high load state, and an acceleration/slope load state. .

In the V-type operation of the wheel loader, the forward travel operation (a), the reverse travel operation (c), the dump operation (e), and the reverse travel boom down operation (f) are determined to be in a low load state, and the excavation operation ( b) may be determined as a heavy load state, and the forward traveling boom-up task (d) may be determined as a high load state. In addition, among the tasks performed by the wheel loader, the slope driving task and the vehicle acceleration task may be determined as an acceleration/slope load state defined between a heavy load state and a high load state.

The above-described embodiments have been described in the case of application to the V-type operation of the wheel loader, but are not limited thereto, and for example, load & carry operation, I-cross operation, etc. It will be appreciated that it can also be applied.

Although the above has been described with reference to the embodiments of the present invention, those skilled in the art can variously modify and change the present invention without departing from the spirit and scope of the present invention described in the claims below. You will understand that you can.

10: wheel loader 12: front body
14: rear body 20: boom
22: boom cylinder 30: bucket
32: bucket cylinder 34: tilt arm
40: cab 50: engine room
100: engine 102: engine speed sensor
110: gear train 120: torque converter
122a, 122b: rotation speed detection sensor 130: transmission
132: vehicle speed detection sensor 140: transmission control device
142: travel pedal 143: travel pedal detection sensor
144: brake pedal 145: brake pedal detection sensor
146: FNR travel lever position detection sensor 150: propeller shaft
152, 154: axle 160: front wheel
162: rear wheel 200: hydraulic pump
202 hydraulic line 204 pressure sensor
210: boom control valve 212: bucket control valve
222: boom cylinder pressure sensor 224: boom angle sensor
234: bucket angle sensor 300: control device
310: work load determination unit 312: signal receiving unit
314: signal selection unit 316: individual load determination unit
318: load state determination unit 320: control signal output unit
330: storage unit 332: first storage unit
334: second storage unit

Claims (20)

Receiving signals indicating work conditions from sensors mounted on the wheel loader when the wheel loader performs a series of tasks;
Selecting a signal necessary for determining each of a plurality of individual loads classified according to loads required in a series of tasks performed by the wheel loader from among the received signals;
Calculating output values representing the plurality of individual load states by performing a pre-learned prediction algorithm on the selected signals;
determining a current workload state by analyzing the calculated output values; and
Outputting a control signal for controlling an engine or transmission of the wheel loader according to the determined current workload state,
Calculating output values indicating whether or not the plurality of individual load conditions are present, respectively,
calculating a first output value indicating whether a low load state is present by performing the prediction algorithm on a first group of signals among the selected signals;
Calculating a second output value indicating whether a heavy load state is present by performing the prediction algorithm on a second group of signals among the selected signals;
calculating a third output value indicating whether a high load state is present by performing the prediction algorithm on a signal of a third group among the selected signals; And
Calculating a fourth output value indicating whether or not an acceleration/slope load state is performed by performing the prediction algorithm on a signal of a fourth group among the selected signals,
Analyzing the calculated output values may include determining that the current work load state is any one of a low load state, a heavy load state, a high load state, and an acceleration/slope load state by analyzing the first to fourth output values. include,
At least one of the boom cylinder pressure signal, the FNR signal, the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal is used as the first group signal and the third group signal, and the hydraulic pump At least one of the main pressure signal, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal is used as a signal of the second group, and at least one of the torque converter speed ratio signal and the accelerator pedal position signal is used as the fourth group signal. A control method for wheel loaders used as a group signal. The method of claim 1, wherein the step of performing the pre-learned prediction algorithm comprises the step of performing a neural network algorithm. delete delete delete The method of claim 1, wherein forward travel, reverse travel, dump work, and reverse boom-down work in the V-shaped operation of the wheel loader are determined to be low load conditions, and excavation work is determined to be heavy load conditions, A control method of a wheel loader, characterized in that the forward traveling boom-up operation is determined as a high load state. delete A signal receiving unit for receiving signals representing work state information from sensors mounted on the wheel loader when the wheel loader performs a series of tasks;
Divide into a plurality of individual load conditions according to the load conditions required by the series of tasks performed by the wheel loader, and select and input signals necessary for determining each of the individual load conditions among the received signals signal selector;
an individual load determination unit calculating an output value representing each of the individual load conditions by performing a pre-learned prediction algorithm on the selected signals;
a load state determining unit analyzing the output values to determine a current work load state; and
Including a control signal output unit for outputting a control signal for controlling the engine or transmission of the wheel loader according to the current work load state of the wheel loader,
The individual load determination unit,
a low load determination circuit unit calculating a first output value indicating whether or not a low load state is present by performing the prediction algorithm on a first group of signals among the selected signals;
a heavy load determination circuit unit calculating a second output value indicating whether a heavy load state is present by performing the prediction algorithm on a second group of signals among the selected signals;
a high load determination circuit unit calculating a third output value indicating whether a high load state is present by performing the prediction algorithm on a third group of signals from among the selected signals; and
An acceleration/slope load determination circuit unit for calculating a fourth output value indicating whether an acceleration/slope load state is present by performing the prediction algorithm on a fourth group of signals among the selected signals;
The load state determining unit analyzes the first to fourth output values to determine that the current work load state is any one of a low load state, a heavy load state, a high load state, and an acceleration/slope load state;
At least one of the boom cylinder pressure signal, the FNR signal, the main pressure signal of the hydraulic pump, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal is used as the first group signal and the third group signal, and the hydraulic pump At least one of the main pressure signal, the vehicle speed signal, the boom position signal, and the torque converter speed ratio signal is used as a signal of the second group, and at least one of the torque converter speed ratio signal and the accelerator pedal position signal is used as the fourth group signal. A control unit on a wheel loader used as a signal for a group. The control device of claim 8, wherein the individual load determination unit includes individual determination circuit units that perform a neural network algorithm. delete delete delete The method of claim 8, wherein the load state determination unit determines that the forward travel, reverse travel, dump operation, and reverse boom down operation in the V-type operation of the wheel loader are low load conditions, and the excavation operation is performed as a heavy load Control device of a wheel loader, characterized in that for determining the state, and determining the forward traveling boom-up operation as a high load state. delete engine;
a working device and traveling device driven by the engine;
a plurality of sensors mounted on the engine, the working device, and the traveling device to detect signals representing work state information of the wheel loader when the wheel loader performs a series of tasks; and
Among the signals received from the sensors, at least one signal capable of determining a plurality of individual load conditions is selected according to a load condition required by a series of tasks performed by the wheel loader, and A control device for determining a current workload state by performing a pre-learned prediction algorithm for each of the individual load states,
The control device,
Selecting signals necessary for determining whether or not the individual load state is present among the received signals;
A pre-learned prediction algorithm is performed on a first group of signals from among the selected signals to calculate a first output value indicating whether or not a low load state is present, and the prediction algorithm is performed on a second group of signals from among the selected signals. to calculate a second output value indicating whether a heavy load state is present, and to calculate a third output value indicating whether a high load state is present by performing the prediction algorithm on a signal of a third group among the selected signals, Performing the prediction algorithm for 4 groups of signals to calculate a fourth output value indicating whether or not there is an acceleration/slope load state,
By analyzing the first to fourth output values, it is determined that the current work load state is any one of a low load state, a heavy load state, a high load state, and an acceleration/slope load state;
A control system of a wheel loader for outputting a control signal for controlling the engine or transmission of the wheel loader according to a current work load state of the wheel loader. delete The control system of claim 15, wherein the control device calculates the output value by performing a neural network algorithm on the selected signals. delete 16. The method of claim 15, wherein the control device determines that forward travel, reverse travel, dump work, and reverse boom down operation in the V-shaped operation of the wheel loader are in a low load state, and the excavation work is a heavy load state. And the control system of the wheel loader, characterized in that for determining the forward traveling boom-up operation as a high load state. delete

Images