Methods for measuring the load weight of a forklift truck

Abstract

Method for measuring the load weight of a forklift truck (2) in which the load rests on a height-adjustable load-bearing device (9) which is raised by a hydraulic lifting device (3), in particular a lifting cylinder, and in which a detection means for the hydraulic pressure (24) of a hydraulic fluid, as well as further detection means for detecting at least one or more of the parameters hydraulic fluid temperature (23), lifting height (h) and hydraulic fluid flow rate are provided, wherein a load weight value is determined by the hydraulic pressure (24) using a correction function based on the detected parameters, characterized in that the correction function represents a multidimensional correction matrix (37) in which pressure correction values ​​for a plurality of operating points of the hydraulic lifting device (3) are stored, and pressure correction values ​​are transmitted to the correction matrix (37) by a data quality monitor (35).if the data quality observer (35) judges the recorded parameters and the hydraulic pressure (24) to be good based on a heuristic assessment.

Description

[0001] The invention relates to a method for measuring the load weight of a forklift truck. In particular, the invention relates to a method for measuring the load weight of a forklift truck in which the load rests on a height-adjustable load-bearing device which is raised by a hydraulic lifting device, in particular a lifting cylinder, and in which sensing means for the hydraulic pressure of a hydraulic fluid, as well as further sensing means for sensing at least one or more of the parameters hydraulic fluid temperature, lifting height and hydraulic fluid flow rate are provided, wherein a load weight value is determined by the hydraulic pressure using a correction function based on the sensed parameters.

[0002] Industrial trucks are equipped with load-handling devices for transporting loads. These devices include a lifting mechanism to, for example, lift a pallet from the ground. Many industrial trucks also offer the option of raising the load-handling device higher on a lifting mast to reach storage locations in racks. Examples of this latter type of industrial truck include forklifts and reach trucks, in contrast to, for example, low-lift pallet trucks, which only have a short lift of fork arms to raise and move a load on a pallet from the ground. Typically, the load fork consists of two tines mounted on a fork carriage, which are used to pick up a pallet.

[0003] For assistance systems, as well as to prevent overloading or unsafe operating conditions, industrial trucks can be equipped with load measuring systems that can detect the weight of a load resting on the load-handling device. Besides direct sensors on the load-handling device, such as strain gauges or sensors that detect the force of weight, for example, pressure sensors, systems that detect the pressure in a lifting cylinder of a hydraulic lifting device and indirectly calculate the load weight from this pressure are particularly well-known for detecting the weight of a load.

[0004] To improve accuracy, it is known to use additional sensors that, for example, detect hydraulic oil flow or the flow rate of a hydraulic fluid and the temperature of the hydraulic oil in order to compensate for dynamic pressure losses in hydraulic lines. It is also known to adapt the load measurement to different operating points of a hydraulic lifting device, for example, by means of a lifting height sensor in a multi-stage lifting mast, in order to take into account different weight proportions of the lifting mast stages and / or effective areas of the lifting cylinders.

[0005] A disadvantage of the described state of the art is its potential for significant inaccuracy when load weight is determined solely via hydraulic oil pressure. This can severely limit the application range of load weight measurement. Extensive correction and consideration of all parameters that could influence the load weight measurement result in considerable parameterization effort and, in particular, require that all these values ​​are available at all times during operation. Furthermore, manufacturing the industrial truck becomes more complex, as a system for determining load weight must be tested and calibrated at all operating points or conditions, or at least at a large number of them, in order to determine the appropriate correction factors.

[0006] A further disadvantage is that some methods for determining load weight are limited to specific operating points, for example, only the first lifting stage of a multi-stage lifting mast or a specific temperature range of the hydraulic oil, such as a hydraulic oil temperature of 85 °C. This significantly limits the applicability of load weight measurement.

[0007] It is also known to improve the measurement accuracy of load weight by using a special measurement procedure, for example, by first lifting a load upwards over a defined distance and then moving it downwards. This can compensate for frictional effects that arise from a shift in the symmetry between upward and downward movement. Such a shift typically results from asymmetries in the hydraulic seals, particularly those of lifting cylinders.

[0008] However, a disadvantage of this well-known method is that it is relatively time-consuming and reduces the handling capacity of the industrial truck.

[0009] For example, EP 1 953 114 B1 discloses a method for measuring the weight of a load, taking into account the properties of the hydraulic fluid, its temperature and volume flow.

[0010] However, a disadvantage of this state of the art is that only an average value for the characteristic parameter of friction is taken into account, and therefore, particularly with very small, very slow movements or at the moment of starting a lifting movement, large inaccuracies and fluctuations result.

[0011] The present invention is based on the objective of providing a method for load measurement in a forklift truck that avoids the aforementioned disadvantages and enables a quick and accurate detection of a load resting on a load-bearing device in as many operating situations as possible.

[0012] This problem is solved by a method having the features of independent claim 1. Advantageous embodiments of the invention are specified in the dependent claims.

[0013] The object of the invention is achieved by a method for measuring the load weight of a forklift truck in which the load rests on a height-adjustable load-bearing device which is raised by a hydraulic lifting device, in particular a lifting cylinder, and in which a detection means for the hydraulic pressure of a hydraulic fluid, as well as further detection means for detecting at least one or more of the parameters hydraulic fluid temperature, lifting height and hydraulic fluid flow rate are provided, wherein a load weight value is determined by the hydraulic pressure using a correction function based on the detected parameters, the correction function being a multidimensional correction matrix in which pressure correction values ​​for a plurality of operating points of the hydraulic lifting device are stored, and pressure correction values ​​are transmitted to the correction matrix by a data quality observer.if the data quality observer, based on a heuristic assessment, judges the recorded parameters and the hydraulic pressure to be good.

[0014] The load weight can be determined using a linear equation by calculating a proportional value to the hydraulic pressure value plus an offset value, which can be determined, for example, by a stroke stage, friction errors, a dynamic pressure drop, or other effects. The coefficients of this linear equation are determined by the pressure correction values ​​of the correction matrix. In addition to the hydraulic pressure, the data quality monitor initially receives one or more of the following parameters as input data: hydraulic fluid temperature, stroke height, and hydraulic fluid flow rate. Other possible parameters include the deflection or flow rates of the hydraulic valves connected to a joystick or hand lever control of the hydraulic system, vehicle speed, the steering angle of the industrial truck, and even, for example, the rotational speed of a hydraulic pump. The data quality monitor then heuristically analyzes the quality of the data in a model created by the data quality monitor.The quality of the recorded parameter data or measurement data is inferred. This involves making a probable statement based on a limited data set, rather than a complete analysis. Evaluating data quality ensures, for example, that only data or pressure correction values ​​rated as sufficiently good are stored in the correction matrix. This filtering process prevents the system from becoming overloaded with data or from data originating from measurements and acquisitions of poor quality. Alternatively, a purely analytical evaluation can be performed instead of a heuristic one.

[0015] Due to continuous error compensation, the system is highly accurate, and the pressure correction matrix captures all operating points in a multidimensional space. This eliminates the dependency of the load weight determination on a specific operating point. The system also possesses self-learning capabilities, eliminating the need for complex parameterization during the manufacturing or production of the industrial truck. For example, an initial set of pressure correction values, which need not encompass all values ​​in the correction matrix, could be stored in the software during manufacturing. Various criteria, such as heuristic evaluation, are conceivable for performance assessment.In a hydraulic lifting device, an oscillating hydraulic oil column can occur, leading to interference due to wave propagation and thus errors in pressure measurement. The degree of interference is usually not easily measurable with simple measuring instruments. For example, an experimentally determined settling time can be used, as the oscillations can be assumed to have completely subsided after a sufficiently long period. The hydraulic oil column is excited by accelerations in the hydraulic oil flow, such as a jerk. These accelerations can be caused by stroke height accelerations, which can be determined directly as a parameter or indirectly by the opening or changes in the valve position of a hydraulic valve, for example, by the movement of a joystick in a control device.Furthermore, vehicle movement on uneven ground can cause vibrations in a lifting device, such as a lifting mast, and thus in the hydraulic oil column. This can be detected by recording travel speeds as a parameter. It is also possible to use a suitable algorithm to account for the aging of the hydraulic system or the industrial truck, thereby incorporating factors such as its maintenance status.

[0016] The lifting height as a parameter can be determined, for example, by measuring the distance between a load-handling device and a reference position using a sensor. Hydraulic fluid temperature, or oil temperature, can be determined by evaluating a temperature sensor immersed in hydraulic fluid. Another parameter that can be used is the hydraulic fluid flow rate, which can be determined, for example, by measuring the deflection of a hand-operated valve or the coil current of an electric hydraulic valve, such as one controlled by a joystick. The actual hydraulic fluid flow rate can potentially be determined using a stored table and pressure measurements.

[0017] In an advantageous embodiment of the method, statistical information on the quality of the collected data is stored in one or more delta matrices with dimensions and working space identical to the correction matrix, in particular values ​​on the dispersion of the print correction values ​​and / or the data history of the print correction values.

[0018] In the actual correction matrix, pressure correction values ​​for individual operating points are stored as a function of several parameters or at least one parameter. This could be, for example, a delta value for the pressure or a correction factor for the load weight, depending on the stroke height, hydraulic oil temperature, stroke speed, load, and load distance. Statistical information about the quality of the collected data in the correction matrix is ​​stored in one or more separate, additional delta matrices. Such a delta matrix has the same dimensions and working space as the correction matrix. For example, it contains a maximum of values ​​for the same operating points. The statistical information can include, for example, the dispersion of the correction data. The size of the data history can also be stored. Based on this, correction data values ​​can be determined.Entries in the correction matrix with a correspondingly short data history, i.e., a small number of previously stored values, and / or large dispersion, are then assessed as being of low quality and processed accordingly.

[0019] Advantageously, the data quality observer stores a last good value for the load weight value and, in the event that the recorded parameters and the hydraulic pressure are rated as poor due to a further heuristic assessment or the same heuristic assessment, this stored load weight value is used as the load weight value.

[0020] The data quality monitor thus prevents both the calculation of a load weight value based on parameter values ​​rated as poor and the use of such parameter values ​​for populating the correction matrix. The evaluation algorithms, which can be either heuristic or analytical, can be identical to those used to populate the correction matrix. Alternatively, however, different criteria can be applied, particularly those with a different level of stringency.

[0021] In a further development of the procedure, an interpolator is used to generate intermediate values ​​for the pressure correction values ​​between the operating points.

[0022] Due to its numerical finiteness or limitations, the correction matrix must operate in a classified manner and can only store values ​​for specific operating points. Furthermore, the correction matrix may only be populated at runtime, for example, if only a minimal set of pressure correction values ​​is initially stored. The interpolator performs interpolation for intermediate values ​​between two stored pressure correction values ​​for two operating points. Even if, for example, a specific parameter is not included in the dimensionality of the correction matrix, the interpolator can incorporate it, for instance, through an analytical model that performs the corresponding calculation.

[0023] The recorded parameters and the hydraulic pressure can be passed through a filter, in particular a low-pass filter, before a load weight value is calculated from them.

[0024] A general filter unit, implemented as a software module, serves as the filter. In the simplest case, such a general filter unit is a low-pass filter for smoothing the input data.

[0025] In a favorable embodiment of the procedure, a load change observer assesses, based on a heuristic evaluation of the recorded parameters and the hydraulic pressure, whether the load weight value may have changed.

[0026] Based on input data such as vehicle speed, hydraulic pressure, lifting height, and lifting speed, as well as tilt pressure if a tilting device is provided for the load-handling attachment, the load change monitor checks whether the load weight on the load-handling attachment has changed. This is done through a heuristic evaluation or heuristic algorithms applied to the input parameters. An analytical evaluation is also conceivable. In addition to the parameters mentioned, other parameters can also be used.

[0027] It becomes advantageous if the load weight value may have changed, and no pressure correction values ​​are transmitted from the data quality observer to the correction matrix.

[0028] This prevents the accumulation of pressure correction values ​​for mixed load weights. For example, when the load changes, the last data stored in buffers or similar is discarded and not taken into account.

[0029] The print correction values ​​can be fed into the correction matrix via a data collector.

[0030] Advantageously, the data collector does not receive the pressure correction values ​​from the data quality observer if the load change observer indicates a change in the load weight value.

[0031] The data collector receives the parameter values ​​for determining the pressure correction values, or the pressure correction values ​​that have been checked and approved by the data quality observer, as well as information from the load change observer. The data collector then operates according to the specifications of the data quality observer and collects pressure correction values ​​depending on the dimensions of the working space and the parameters listed previously and described below. In an ideal system, friction would behave symmetrically when collecting pressure correction values. However, in a real system, a shift in symmetry generally occurs, which can be determined experimentally with sufficient accuracy, for example, and is reflected in the pressure correction values. This shift is typically caused, for instance, by asymmetries in the hydraulic seals in the system described here.

[0032] An evaluation and display unit can receive data from the data quality observer and determine and display a value for the quality of the load weight value.

[0033] The evaluation unit does not necessarily have to be connected to a display unit; for example, such a quality value can also be displayed by a third device or used solely within a control system. Heuristic and statistical algorithms are used to evaluate the currently determined load weight value. The resulting quality value is then displayed to the operator. A suitable format for this could be, for example, a bar graph.

[0034] Advantageously, the evaluation and display unit receives data from the load change observer and / or the interpolator.

[0035] A large distance to a support point of a pressure correction value of the correction matrix during interpolation in the interpolator is potentially associated with poor accuracy and can be evaluated accordingly.

[0036] The evaluation and display unit can receive data from the correction matrix and assess the quality based on the distance to the print correction value.

[0037] The evaluation and display unit can receive data from one or more delta matrices and use this data to assess the quality.

[0038] This takes into account the statistical information regarding the quality of the collected data.

[0039] Advantageously, the recorded parameters can include lifting speed and / or lifting acceleration and / or hydraulic pump torque and / or hydraulic pump speed and / or travel speed of the industrial truck and / or a steering angle of the industrial truck and / or valve openings of hydraulic control devices and / or a load distance.

[0040] The lifting speed and lifting acceleration can be determined from the first and second derivatives of the lifting height. The vehicle speed can be determined from the right and left wheel rotation speeds and the corresponding tire diameters, whereby a steering angle can be determined, for example, not only by direct measurement but also by calculating the difference between the wheel rotation speeds.

[0041] If the industrial truck has a tilting device for a load handling device, the parameters may include a hydraulic pressure in a tilting device and / or a tilting angle.

[0042] The tilt pressure can be determined, for example, via a sensor, whereby when using differential cylinders an equivalent pressure is determined via an area-weighted differential pressure measurement.

[0043] The industrial truck can include a lifting mast, in particular a lifting mast with multiple lifting stages, and the extended lifting stage can be recorded as a parameter.

[0044] Different lifting stages can result in different effective cylinder areas, for example, in a lifting cylinder used as a lifting device. This can be taken into account as a parameter or in the pressure correction value. Furthermore, with each extended lifting stage of a multi-stage lifting mast, the unladen weight of the mast components, which is also lifted, changes and also affects the hydraulic pressure.

[0045] In a measurement data calculation module, implemented for example as software, the hydraulic pressure values ​​are converted into a load weight value using interpolated coefficients and a linear equation. The load weight value can be displayed as a mass on a screen. The last valid value can be used for this purpose. However, a value can also be determined, for example, by filtering a measurement history, or no value can be displayed if one cannot be determined based on the available data.

[0046] Further advantages and details of the invention are explained in more detail with reference to the exemplary embodiments shown in the schematic figures. Here, Fig. 1 schematically a forklift truck as an example of a material handling vehicle in which the method according to the invention is used, and Fig. 2 schematically in a diagram the method according to the invention.

[0047] The Fig. Figure 1 schematically shows a forklift 1 as an example of a material handling vehicle 2 in which the method according to the invention is used. The forklift 1 has a hydraulic lifting device 3, which includes a hydraulic pump 4. The hydraulic fluid pumped by the pump is supplied to a lifting cylinder via hydraulic valves 5, controlled in this example by a joystick 6. The hydraulic pump 4 draws the hydraulic fluid from a hydraulic tank 7 with a hydraulic oil temperature sensor 8. The hydraulic lifting device 3 raises a load-handling device 9 on a lifting mast 10 together with a load weight 13. The lifting mast 10 has several lifting stages in the form of sequentially extending mast sections, and a lifting stage sensor 11 detects the extension of each subsequent stage. Further sensors detect a tilt angle β of the lifting mast 10 with the load-handling device 9 and a lifting height h of the load-handling device 9.By recording the wheel speeds of the front drive wheels 12 of the forklift 1, a steering angle α can be determined based on the speed differences between the sides and by taking into account the tire diameter.

[0048] The Fig.Figure 2 schematically illustrates the method according to the invention in a diagram. A sensor data processor 14 receives values ​​from a lift height sensor 15, a temperature sensor 8, a hydraulic pressure sensor 16, a left wheel speed sensor 17, a right wheel speed sensor 18, signal values ​​from the joystick 6, values ​​from a mast tilt sensor 19, and a tilt pressure sensor 20. The sensor data processor 14 uses these values ​​to generate a lift height h, a lift speed 21, a lift acceleration 22, a hydraulic oil temperature 23, a hydraulic pressure 24, a hydraulic pump torque 25, a hydraulic pump speed 26, a mast tilt 27, a vehicle speed 28, the steering angle α, a tilt pressure 30, a load distance 31, and a valve opening value 32. These parameter values ​​and the hydraulic pressure 24 are passed via a filter 33 to a load change observer 34 and a data quality observer 35 as well as a data collector 36.Depending on the results of the data quality monitor 35, the data collector 36 writes pressure correction values ​​into a correction matrix 37. Furthermore, the data collector 36 populates two delta matrices 38 and 39 with statistical information on the quality of the collected data. The parameter values ​​processed by the filter 33 are fed to a load calculation 40, which determines a load weight value for a display 41. For this purpose, a matrix manager 42 forwards data correction values ​​to an interpolator 43. An evaluation and display unit 44 receives data from the matrix manager 42, the data quality monitor 35, the load change monitor 34, and the interpolator 43.

Claims

[1] Method for measuring the load weight of a forklift truck (2) in which the load rests on a height-adjustable load-handling device (9) which is raised by a hydraulic lifting device (3), in particular a lifting cylinder, and in which a detection means for the hydraulic pressure (24) of a hydraulic fluid, as well as further detection means for detecting at least one or more of the parameters hydraulic fluid temperature (23), lifting height (h) and hydraulic fluid flow rate are provided, wherein a load weight value is determined by the hydraulic pressure (24) using a correction function based on the detected parameters, characterized by, that the correction function represents a multidimensional correction matrix (37) in which pressure correction values ​​for a plurality of operating points of the hydraulic lifting device (3) are stored, and pressure correction values ​​are transmitted to the correction matrix (37) by a data quality observer (35) when the data quality observer (35) judges the recorded parameters and the hydraulic pressure (24) to be good based on a heuristic assessment. [2] Method according to claim 1, characterized by , that in one or more delta matrices (38,39) with dimension and workspace identical to the correction matrix (37) statistical information on the quality of the collected data is stored, in particular values ​​on the dispersion of the print correction values ​​and / or the data history of the print correction values. [3] Method according to claim 1 or 2, characterized by, that the data quality observer (35) stores a last good value for the load weight value and in the event that the recorded parameters and the hydraulic pressure (24) are rated as poor due to a further heuristic assessment or the same heuristic assessment, this stored load weight value is taken as the load weight value. [4] Method according to any one of claims 1 to 3, characterized by , that intermediate values ​​for the pressure correction values ​​between the operating points are formed by an interpolator (43). [5] Method according to any one of claims 1 to 4, characterized by , that the recorded parameters and the hydraulic pressure (24) are passed through a filter (33), in particular a low-pass filter, before a load weight value is calculated from them. [6] Method according to any one of claims 1 to 5, characterized by, that a load change observer (34) assesses, based on a heuristic evaluation of the recorded parameters and the hydraulic pressure (24), whether the load weight value may have changed. [7] Method according to claim 6, characterized by , that if the load weight value may have changed, no pressure correction values ​​are transmitted from the data quality observer (35) to the correction matrix (37). [8] Method according to any one of claims 1 to 7, characterized by , that the print correction values ​​are supplied to the correction matrix (37) via a data collector (36). [9] Method according to claim 8 and claim 7, characterized by , that the data collector (36) does not receive the pressure correction values ​​from the data quality observer (35) when the load change observer (34) indicates a change in the load weight value. [10] Method according to any one of claims 1 to 9, characterized by, that an evaluation and display unit (44) receives data from the data quality observer (35) and determines and displays a value for the quality of the load weight value. [11] Method according to claim 10, characterized by , that the evaluation and display unit (44) receives data from the load change observer (34) and / or the interpolator (43). [12] Method according to claim 11, characterized by , that the evaluation and display unit (44) receives data from the correction matrix (37) and evaluates the quality based on the distance to the print correction value. [13] Method according to one of claims 10 to 12 in conjunction with claim 2, characterized by , that the evaluation and display unit (44) receives data from one or more delta matrices (38,39) and uses these to evaluate the quality. [14] Method according to any one of claims 1 to 13, characterized by, that the recorded parameters may include lifting speed (21) and / or lifting acceleration (22) and / or hydraulic pump torque (25) and / or hydraulic pump speed (26) and / or travel speed (28) of the industrial truck (2) and / or a steering angle (α) of the industrial truck (2) and / or valve openings (32) of hydraulic control devices and / or a load distance (31). [15] Method according to any one of claims 1 to 14, characterized by , that the industrial truck (2) has a tilting device for a load handling device (9) and the parameters may include a hydraulic pressure (30) in a tilting device and / or a tilting angle (β). [16] Method according to any one of claims 1 to 15, characterized by , that the industrial truck (2) comprises a lifting mast (10), in particular a lifting mast (19) with several lifting stages, and the extended lifting stage is recorded as a parameter.

Images