Method for controlling a working machine system for work operation

Abstract

The invention relates to a method for controlling a working machine system (1) during operation, wherein the generator (9) for the electrical supply of the working units (3) is driven by the combustion engine (10) via a clutch (11). It is proposed that during operation, the system control (14) determines the actual load (B) of the clutch (11) in the time domain using a measurement routine and derives an actual load spectrum (b) of the clutch (11) over a measurement frequency range. The system control (14) then uses the actual load spectrum (b) of the clutch (11) and a predetermined limit load spectrum (G) of the clutch (11), which represents a nominal limit load state of the clutch (11), to determine the actual load state (B) of the clutch (11) with respect to at least one optimization criterion relating to the actual utilization.

Description

[0001] The present invention relates to a method for controlling a work machine system for work operation according to the preamble of claim 1, as well as a system control of a work machine system as such according to claim 15, and a work machine system as such according to claim 16.

[0002] The machine system in question, to which the method can be applied, comprises a number of electrical work units whose energy is generated, at least temporarily, by an electric generator driven by an internal combustion engine. Such machines are found in numerous fields of application, such as agriculture, construction, and transportation. The machine system can be self-propelled or non-self-propelled, for example, stationary or towed like a trailer.

[0003] The prior art (US 10,968,842 B2), from which the invention is based, relates to a method for controlling a machine system with a series of working units, in which, as above, a generator is driven by an internal combustion engine to provide the electrical energy for the drive units. To detect wear of the internal combustion engine and other components, a generator parameter such as the generator current is measured in the time domain and evaluated in the frequency domain. Appropriate maintenance work is then initiated.

[0004] The established method allows for a rapid response to existing wear, for example, by replacing defective parts in a very short time. However, it cannot reverse wear that has already occurred. Preventing such wear is limited to mechanical measures, such as wear-reducing designs, particularly through the use of flywheels. These measures are expensive and energy-intensive during operation.

[0005] The invention is based on the problem of designing and further developing the known method in such a way that the probability of wear on the functional part of the coupling between combustion engine and coupling is actively reduced in a simple and cost-effective manner, thus increasing the service life.

[0006] The above problem is solved by the features of the characterizing part of claim 1.

[0007] The proposed design assumes that the machine system comprises multiple electrical work units and a power supply arrangement for these units, the latter consisting of an electric generator and an internal combustion engine. The generator is driven by the internal combustion engine via a clutch to supply electrical power to the work units. The machine system includes an electronic control unit for controlling the work units. Depending on the current draw by the work units, a fluctuating load occurs between the internal combustion engine and the generator. This results in a fluctuating, and depending on the work unit, a pulsating or oscillating load on the clutch, which is always associated with a certain degree of wear.

[0008] The fundamental consideration is that clutch wear is influenced not only by the clutch's operating time, but primarily by the load state and, in particular, the dynamics of the alternating loads on the clutch. Frequency-dependent variations in load are tolerable without excessive wear. Given that a machine system can be operated differently for the same task, possibly with temporarily reduced operating speeds, the basic idea is to optimize the actual load spectrum with a view to reducing wear in an optimization routine.

[0009] The proposed optimization routine focuses on the actual load capacity of the coupling. In this context, the "actual load capacity of the coupling" indicates how far the current load state deviates from a nominal limit load state. The nominal limit load state, defined before the coupling is actually operated, is the permissible load state that generates no significant or only minimal wear. The nominal limit load state can be determined for a coupling through testing and is often specified by the coupling manufacturer. A limit load spectrum in the frequency domain is associated with the limit load state, as the magnitude of the permissible load changes with frequency, as mentioned above. Accordingly, the load capacity can be represented by a load spectrum or by a resulting load value.

[0010] The optimization routine now focuses on fulfilling an optimization criterion related to utilization. Usually, the primary goal is to reduce utilization across the entire measurement frequency range. However, it's also conceivable that the optimization routine is based on several conflicting optimization criteria, ultimately aiming to maintain a moderate utilization level for different frequencies.

[0011] Specifically, it is proposed that during operation, the system control determines the actual load capacity of the clutch with respect to the limit load capacity from the actual load spectrum of the clutch and a predetermined limit load spectrum of the clutch, which represents a nominal limit load state of the clutch, according to a calculation rule, and that an optimization routine is carried out to influence the actual load state of the clutch with respect to at least one optimization criterion relating to the actual load capacity.

[0012] As a result, the probability of clutch wear can be reduced in a particularly simple and cost-effective way, namely purely through control engineering. It is only necessary to run the proposed optimization routine to optimize and, in particular, reduce the actual load. Mechanical measures to reduce wear are not required.

[0013] Of particular importance here is that the optimization of capacity utilization takes place during operation. The optimization is therefore not performed offline, separately from operations, but online while the machine is running. Preferably, the optimization runs in real time, i.e., with a reproducible response time.

[0014] The limit load spectrum can, in principle, be stored in a system memory of the control unit or in a system memory of another controller and read from there to be used for the proposed optimization routine. This demonstrates a high degree of flexibility. Accordingly, the limit load spectrum can also be adjusted if relevant insights into wear are gained from the operation of the clutch (claim 2).

[0015] According to claim 3, the actual load is the torque transmitted by the coupling, which can be determined in a particularly simple way via the current values ​​of the generator.

[0016] According to claim 4, in a particularly preferred embodiment, the frequency analysis is a discrete spectral analysis applied cyclically over a discrete time interval, more preferably a discrete Fourier analysis and more preferably a Fast Fourier Transform, which can be realized, for example, using a simple ring buffer or First-In-First-Out memory.

[0017] The further preferred embodiment according to claim 5 proposes the calculation of an actual utilization spectrum, which reveals in which partial frequency range a problematic increase in the actual utilization occurs. As will be explained below, on this basis, those working units that generate a load on the clutch in precisely this partial frequency range can easily be subjected to modified control.

[0018] The further preferred embodiment according to claim 6 relates to an advantageous optimization criterion according to which the limiting utilization spectrum should be undercut by the actual utilization spectrum. This optimization criterion can be limited to a partial frequency range, so that advantageously another optimization criterion can be directed to a further partial frequency range. The term "partial frequency range" here consistently means that the partial frequency range is a frequency interval within the measurement frequency range.

[0019] In the further preferred embodiment according to claim 7, an optimization criterion is sought according to which a predetermined amplitude distribution and / or envelope of the actual load spectrum and / or the actual utilization spectrum is to be maintained, such that it is possible, for example, to fulfill the optimization criterion without the actual load spectrum falling below the limit load spectrum. This results in particularly high flexibility in defining the optimization routine.

[0020] Alternatively or additionally to the actual utilization spectrum, an actual utilization value can be calculated according to claim 8, which summarizes the utilization across the measurement frequency range. As a one-dimensional variable, the actual utilization value can be used for the optimization routine with minimal computational effort.

[0021] A particularly effective definition of the actual utilization value is the subject of claim 9. The frequency-wise summation of the values ​​of the actual utilization spectrum can also be performed with low computational intensity.

[0022] The optimization criterion defined in claim 10 is based on comparing the actual utilization value with a limit utilization value, which is generally not computationally intensive and therefore particularly suitable for optimization in real-time operation or near real-time operation.

[0023] The preferred embodiments according to claims 11 to 15 relate to preferred variants for influencing the actual load state of the clutch within the framework of the optimization routine. In the particularly preferred embodiment according to claim 11, a change in the control of at least one of the working units is provided. This can, for example, mean the targeted switching on and off of working units (claim 12). Alternatively or additionally, it can also mean an intervention in the control parameters of the control of at least one working unit (claim 13). The invention places virtually no limits on the freedom to influence the actual load state.

[0024] If the system control provides information via a load model about which work unit contributes what amount to the actual load spectrum, optimization can be performed particularly efficiently (claim 14). This makes it possible to influence the actual load spectrum in a targeted manner by intervening in the control of a specifically selected work unit. It can be expected that the optimization routine can thus run with minimal changes to the control while simultaneously achieving a particularly good optimization result.

[0025] According to a further teaching as described in claim 15, which has independent significance, a system control for a machine system for carrying out a proposed method is claimed. Reference may be made to all descriptions of the proposed method.

[0026] According to a further teaching as claimed in claim 16, which also has independent significance, a machine system for carrying out a proposed method is claimed. Reference may also be made in this respect to all descriptions of the proposed method.

[0027] The invention will now be explained in more detail with reference to a drawing that merely illustrates exemplary embodiments. The drawing shows Fig. 1. A proposed working machine system for carrying out a proposed procedure in its general functional structure and Fig. 2 a) the actual load in the time domain over several time periods and b) the derivation of the actual load spectrum for two time periods from representation a) as well as c) the actual utilization over the time periods from representation a).

[0028] The method in question relates to a method for controlling a work machine system 1 for operation. The work machine system 1 comprises the mechanical work machine 2 and further control-related components, which will be explained later. As explained in the introductory part of the description, the application areas of the proposed work machine system 1 are diverse. In the illustrated embodiment, the work machine system 1 is a crawler excavator for the construction industry. In another, particularly preferred embodiment, the work machine system 1 is an agricultural work machine system 1, in particular a self-propelled agricultural work machine system 1. In this sense, the work machine system 1 can be a tractor or a harvesting machine, for example, a combine harvester.However, it is also conceivable that the working machine system 1 is not designed to be self-propelled, as was also explained in the introductory part of the description.

[0029] The in Fig. The working machine system 1 shown in Figure 1 has a plurality of electrical working units 3. These working units 3 are, preferably, a first electric drive 4, a second electric drive 5, a first hydraulic pump 6, and a second hydraulic pump 7. The hydraulic pumps preferably supply hydraulic motors and are generally referred to as "auxiliary units". Other working units 3 are conceivable.

[0030] The machine system 1 is further equipped with a power supply arrangement 8 for the working units 3. The power supply arrangement 8 comprises an electric generator 9 and an internal combustion engine 10, wherein the generator 9 is driven by the internal combustion engine 10 to supply the working units 3 with electricity via a coupling 11, which is only indicated in the drawing. The coupling 11 could, for example, be an elastomer coupling.

[0031] In principle, an electrical energy storage device such as a battery or a supercapacitor can be used as a buffer. However, in a particularly preferred embodiment, such a buffer is omitted. In the latter case, any mechanical load fluctuation on the generator side is transmitted largely undamped to the generator 9 and thus to the drive train between the combustion engine 10 and the generator 9. This means that any load fluctuation on the generator side, and thus any frequency component in the load fluctuation on the generator side, causes a corresponding load fluctuation in the coupling 11. Here, the proposed solution is particularly beneficial in order to largely avoid premature wear.

[0032] An AC-DC converter 12 is connected downstream of the generator 9, providing the supply voltage for the working units 3. Each working unit 3 is preferably connected upstream of a DC-AC converter 13, which inverts the DC voltage into the required AC voltage.

[0033] Finally, the working machine system 1 is equipped with an electronic control system 14 for controlling the working units 3. The control system 14 can, as shown in Fig. Figure 1 shows that the system control could be designed as a central control unit. However, it is also conceivable that the system control unit 14 comprises a number of decentrally distributed control units. Some of these control units could even be spatially separated from the machine 2 itself and, for example, be cloud-based.

[0034] The working machine system 1 in this case is therefore a system that is at least partially powered by electrical energy. Here, and preferably, the working machine system 1 is powered entirely electrically. However, a working machine system 1 that is only partially powered by electrical energy is also conceivable, which has a direct mechanical drive train to the combustion engine 10 for driving certain working units.

[0035] The essential point is that, during operation, the system control 14 determines the actual load B of the clutch 11 in the time domain using a measurement routine. The term "actual" always means that it refers to the current value of the respective quantity.

[0036] The actual load B, here the torque, is measured in the time domain, preferably by measuring the generator current 9, which is proportional to the load on the coupling 11, here the torque transmitted by the coupling 11. The actual load B in the time domain shows Fig. 2a).

[0037] From the actual load B, an actual load spectrum b of the coupling 11 is derived over a measurement frequency range. The term "actual" refers to the current load spectrum, in this case, the current torque spectrum. The term "current" is to be understood in a broader sense here, given that calculating a load spectrum always requires a certain time interval 15, as will be explained later. Fig. 2b) shows the actual load spectrum b for two consecutive time periods 15.

[0038] At the in Fig. In the embodiment shown in Figure 2, which is preferred in this respect, the time intervals 15 are consecutive. However, it is also conceivable that the time intervals 15 overlap.

[0039] Fig. Figure 2b) also shows a limiting stress spectrum G. A fine dashed line is shown above and below the limiting stress spectrum G. This is intended to highlight that the limiting stress spectrum G is selected from a family of curves depending on certain environmental conditions, especially temperature.

[0040] For low wear, the limiting load spectrum G must be undercut by the actual load spectrum b. This is shown in the left graph of Fig. 2b) the case, but not in the right-hand graph of Fig. 2b), in which two load peaks P exceed the limiting load spectrum G.

[0041] The system control 14 now uses the actual load spectrum b of the coupling 11 and a predetermined limit load spectrum G of the coupling 11, which represents a nominal limit load state of the coupling 11, to determine an actual utilization of the coupling 11 with respect to the limit load state according to a calculation rule. The actual utilization is an indicator of the difference between the actual load spectrum b and the limit load spectrum G and serves as the basis for an optimization routine.

[0042] The optimization routine serves to influence the actual load state B of the coupling 11 with regard to at least one optimization criterion concerning the actual utilization. The aim here is to influence the actual load state B based on the actual utilization in order to fulfill the respective optimization criterion(s). For example, the optimization criterion is simply aimed at minimizing the utilization. Other preferred optimization criteria are specified below.

[0043] Unlike the actual load spectrum b, the limit load spectrum G is not generated and stored online in real time, but offline. As mentioned above, the limit load spectrum G is often based on long-term tests regarding wear phenomena. It is frequently issued by the manufacturer of coupling 11 as a usage recommendation.

[0044] Preferably, the limit load spectrum G is stored in a system memory of the control unit 14. Then, preferably, a spectrum from the system memory of the control unit 14 is read out as the limit load spectrum G to determine the limit load spectrum G. Alternatively, the limit load spectrum G can also be stored in a secondary control unit 16, for example, a cloud-based control unit, and read out from there.

[0045] It may be necessary to select the correct limiting load spectrum G from a set of stored limiting load spectra, depending on various environmental conditions. Preferably, one of these environmental conditions is the coupling operating temperature. In this case, to determine the limiting load spectrum G, the coupling operating temperature is first determined, after which a spectrum is read from the relevant system memory as the limiting load spectrum G, based on the coupling operating temperature.

[0046] Numerous advantageous variants are conceivable for defining the load on the coupling 11. Here, and preferably, the actual load B of the coupling 11 is the torque transmitted by the coupling 11. This load can also be determined in different ways. In a preferred variant, the current values ​​of the generator 9, which represent the torque transmitted by the coupling 11, are determined over time to ascertain the actual load B. Alternatively or additionally, the torque signals from a torque sensor assigned to the coupling 11 (not shown here) are received to determine the actual load B.

[0047] The actual load spectrum b is preferably obtained from a frequency analysis applied to the actual load B in the time domain. Preferably, a discrete spectral analysis, more preferably a discrete Fourier analysis, and more preferably a fast Fourier transform are used, each of which is computationally inexpensive.

[0048] Specifically, the actual load B is first determined cyclically over a discrete time interval 15 and temporarily stored. This is in Fig. 2a) shown. In principle, an acyclic determination of the actual load B in the time domain is also conceivable. Intermediate storage is performed in a ring buffer or similar device. In this process, the actual load B may be preprocessed, for example by averaging or other filtering and / or by compensating for mass effects, in particular the moment of inertia of generator 9. The frequency analysis is then applied to the temporarily stored and preferably preprocessed, in particular averaging-adjusted, actual load B.

[0049] The discrete time interval 15, over which the actual load B is determined, can be of varying lengths. Preferably, the length of the time interval 15 is chosen to be between 0.5 and 3 s, preferably 1 s. In the latter case, a new, i.e., current, actual load spectrum b is available every second.

[0050] Furthermore, and preferably, it is provided here that an actual utilization spectrum is calculated according to the calculation rule for determining the actual utilization. Thus, the actual utilization is obtained according to the calculation rule by the computational modification of the actual utilization spectrum b based on the limiting utilization spectrum G. Preferably, the actual utilization is obtained according to the calculation rule by the frequency-wise application of computational operations to the actual utilization spectrum b based on the limiting utilization spectrum. Specifically, this preferably means that the actual utilization is obtained by the frequency-wise relation of the actual utilization spectrum b to the limiting utilization spectrum G. The actual utilization spectrum determined in this way provides frequency-wise information about how far the actual utilization spectrum b is from the limiting utilization spectrum G.in which partial frequency ranges the actual load spectrum b exceeds the limit load spectrum G.

[0051] Accordingly, it is preferably proposed that the optimization criterion be the actual load spectrum b remaining below the limiting load spectrum G over at least one partial frequency range. Alternatively or additionally, it can be provided that the optimization criterion is the actual load spectrum remaining below the limiting utilization spectrum over at least one partial frequency range. In both cases, the optimization routine is used to influence the actual load state B such that the actual load spectrum b remains below the limiting load spectrum G. The actual utilization spectrum preferably provides the basis for this.

[0052] However, it is also possible that an optimization criterion focuses less on the magnitude of the actual load values ​​B, but rather on the distribution of the actual load values ​​B in the frequency domain. In this case, it is preferable that an optimization criterion is adherence to a predetermined amplitude distribution and / or envelope of the actual load spectrum b and / or the actual utilization spectrum over at least a partial frequency range.

[0053] A particularly simple calculation method involves calculating an actual utilization value A according to the formula for determining actual utilization. This value is derived from a correlation between the actual utilization spectrum b and the limit utilization spectrum G. The term "correlation" is to be interpreted broadly here and encompasses any calculation of the actual utilization spectrum b and the limit utilization spectrum G to produce a resulting numerical value.

[0054] In a particularly preferred embodiment, it is provided here and preferably that the actual utilization value A is calculated from the actual utilization spectrum, in particular that the actual utilization value A results from a frequency-wise summation of the values ​​of the actual utilization spectrum over the measurement frequency range, possibly weighted according to frequency.

[0055] This translates the actual utilization into an actual utilization value A, which indicates the proximity between the actual utilization spectrum b and the limit utilization spectrum G. The optimization criterion then only involves a simple numerical value, further accelerating the execution of the optimization routine. This is in Fig. 2c) is shown. For each time interval 15 in which the actual load spectrum b is re-determined by frequency analysis, a resulting actual utilization value A is obtained.

[0056] Specifically, the optimization criterion for determining the above actual utilization value A is defined as the actual utilization value A falling below a first threshold utilization value. Multiple threshold utilization values ​​can also be defined, for example, a first threshold utilization value S1 and a second threshold utilization value S2, in order to generate different responses to exceeding the first or second threshold utilization value, respectively. This is also described in Fig. 2c) shown.

[0057] In principle, numerous advantageous variations are conceivable for influencing the actual load state B within the optimization routine. In the simplest case, for example, a warning message is issued to the user via a display 17, informing them that a critical operating point has been reached and requesting them to change the actual load state B. Preferably, the user is shown a recommendation regarding which measures should be taken. For example, the user can be informed which working units 3 are generating the critical actual load values ​​B. Such a recommendation can advantageously be generated based on a load model explained below and displayed via the display 17. The display 17 is preferably located in a driver's cab 18 of the machine 2.

[0058] The optimization routine, and in particular the influencing of the actual load state B, can also run automatically. In this case, it is preferably the case that, depending on the actual load, the system control 14 in the optimization routine causes a change in the control of at least one of the work units 3 in order to trigger the actual load state B in order to achieve the optimization criterion(s).

[0059] The simplest and most effective way to automatically influence the actual load state B is to modify the control by selectively switching working units 3 on and off. This switching on and off is preferably accompanied by a certain time hysteresis to prevent an undesired toggling situation. It can also be advantageous to represent the importance of each working unit 3 for safe operation in a priority model. This priority model indicates the importance of each working unit 3 for safe operation. For example, the first drive unit 4 and the second drive unit 5 are of particular importance for safe operation, while an air conditioning system, as a comfort function, is of secondary importance. In this case, switching off the air conditioning unit 3 is only an option.

[0060] However, it can also be advantageous that the system control 14 includes the control of individual working units 3 and that the change in the control involves an intervention in the control parameters, in particular a reduction of the control dynamics, of the control of at least one working unit 3. This allows the rate of change of the actual load values ​​B to be reduced in a simple way, which accordingly helps to fulfill the optimization criteria aimed at reducing wear in a relatively simple manner.

[0061] The measures for influencing the actual load state B during the optimization routine can be implemented most effectively if a load model is available that maps the frequency-related contribution of individual work units 3 to the actual load spectrum b. Because the system control 14 has information from the load model about which work unit 3 contributes to the actual load state B in which sub-frequency range and to what extent, a load peak detected in a specific sub-frequency range within the actual load spectrum can be selectively eliminated by choosing and controlling, or if necessary, switching off, a suitable work unit 3 that generates a load in precisely that sub-frequency range.

[0062] Specifically, it is preferably intended that the change in the control, in particular the selection of which working unit 3 or units 3 the control is changed for, is based on the load model described above, which represents the frequency-related contribution of individual working units 3 to the actual load spectrum b. The load model can be determined offline based on calculations and / or tests. Alternatively or additionally, the load model can be determined online, i.e., during operation, based on operating and / or sensor data.

[0063] Alternatively or additionally, the optimization routine may also involve modifying the control of the combustion engine 10 to meet the optimization criterion(s). For example, the rotational speed of the combustion engine 10 can be shifted into a range where the loads are reduced, for example, by eliminating resonance effects.

[0064] To further reduce the dynamics in the actual load state B, it may be advantageous to reduce the control dynamics of generator 9. This applies in particular to the voltage regulator of generator 9, which can be set to a reduced dynamic range, accepting additional voltage fluctuations.

[0065] Finally, it should be noted that the proposed procedure within the framework of the optimization routine can also provide guidance for the design of coupling 11 and working units 3, in order to take measures for the best possible implementation of the optimization routine as early as the design phase.

[0066] According to a further teaching, a system control 14 of a working machine system 1 is proposed, which is suitable for carrying out a proposed method. Reference may be made to all details concerning the proposed method and, in particular, to the system control 14.

[0067] According to a further teaching, a working machine system 1 is proposed, which is suitable for carrying out a proposed method. Reference may be made to all details concerning the proposed method and, in particular, the working machine system 1. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 10,968,842 B2

[0003]

Claims

[1] Method for controlling a working machine system (1) for operation, wherein the working machine system (1) comprises a plurality of electrical working units (3) and a power supply arrangement (8) for the working units (3), wherein the power supply arrangement (8) comprises an electric generator (9) and an internal combustion engine (10), wherein the generator (9) is driven by the internal combustion engine (10) via a clutch (11) to supply the working units (3) with electricity, wherein the working machine system (1) comprises an electronic control system (14) for controlling the working units (3), characterized by, that during operation the system control (14) determines the actual load (B) of the clutch (11) in the time domain in a measurement routine and derives an actual load spectrum (b) of the clutch (11) from it via a measurement frequency range, that the system control (14) determines an actual utilization of the clutch (11) with respect to the limit load state according to a calculation rule from the actual load spectrum (b) of the clutch (11) and a predetermined limit load spectrum (G) of the clutch (11), which represents a nominal limit load state of the clutch (11), and that an optimization routine is carried out to influence the actual load state (B) of the clutch (11) with respect to at least one optimization criterion relating to the actual utilization. [2] Method according to claim 1, characterized by, that to determine the limit load spectrum (G) a spectrum is read out from a system memory of the system controller (14) as the limit load spectrum (G) or is read out from the system memory of a secondary controller (16), in particular a cloud-based controller, by means of the system controller (14), and / or that to determine the limit load spectrum (G) the coupling operating temperature is determined and a spectrum is read out from the relevant system memory as the limit load spectrum (G) based on the coupling operating temperature. [3] Method according to claim 1 or 2, characterized by, that the actual load (B) of the coupling (11) is the torque transmitted by the coupling (11), preferably that to determine the actual load (B) the time-dependent current values ​​of the generator (9), which represent as actual load (B) the torque transmitted by the coupling (11), are determined, and / or that to determine the actual load (B) the torque signals are received from a torque sensor associated with the coupling (11). [4] Method according to any one of the preceding claims, characterized by, that to determine the actual load spectrum (b) a frequency analysis, preferably a discrete spectral analysis, further preferably a discrete Fourier analysis, further preferably a Fast Fourier Transform, is applied to the actual load (B) in the time domain, preferably that for this purpose the actual load (B) in the time domain is determined cyclically or acyclically over a discrete time interval (15) and temporarily stored, and that the frequency analysis is applied to the temporarily stored actual load (B). [5] Method according to any one of the preceding claims, characterized by, that according to the calculation rule for calculating the actual utilization, an actual utilization spectrum is calculated, preferably, that according to the calculation rule, the actual utilization results from the computational modification of the actual load spectrum (b) based on the limit load spectrum (G), further preferably, that according to the calculation rule, the actual utilization results from the frequency-wise application of computational operations to the actual load spectrum (b) based on the limit load spectrum, further preferably, that the actual utilization results from the frequency-wise reference of the actual load spectrum (b) to the limit load spectrum (G). [6] Method according to any one of the preceding claims, characterized by , that an optimization criterion is the actual utilization spectrum falling below a predetermined limit utilization spectrum over at least one partial frequency range. [7] Method according to any one of the preceding claims, characterized by , that an optimization criterion is compliance with a predetermined amplitude distribution and / or envelope of the actual load spectrum (b) and / or the actual utilization spectrum over at least a partial frequency range. [8] Method according to any one of the preceding claims, characterized by , that according to the calculation rule for calculating the actual utilization, an actual utilization value (A) is calculated, preferably that the actual utilization value (A) results from a correlation between the actual load spectrum (b) and the limit load spectrum (G) according to the calculation rule. [9] Method according to any one of the preceding claims, characterized by, that the actual utilization value (A) is calculated from the actual utilization spectrum, preferably that the actual utilization value (A) results from a frequency-wise summation of the values ​​of the actual utilization spectrum over the measurement frequency range, possibly weighted according to frequency. [10] Method according to any one of the preceding claims, characterized by , that an optimization criterion is the actual utilization value (A) falling below a limit utilization value. [11] Method according to any one of the preceding claims, characterized by , that in the optimization routine, the system control (14) causes a change in the control of at least one of the working units (3) depending on the actual load. [12] Method according to any one of the preceding claims, characterized by , that the change in control includes the targeted switching on and off of working units (3). [13] Method according to any one of the preceding claims, characterized by , that the system control (14) includes the control of individual working units (3) and that the change in the control includes an intervention in the control parameters, in particular a reduction of the control dynamics, of the control of at least one working unit (3). [14] Method according to any one of the preceding claims, characterized by , that the change in the control, in particular the selection of which working unit (3) or which working units (3) a change in the control is made, is based on a load model that maps the frequency-related contribution of individual working units (3) to the actual load spectrum (b). [15] Control system of a working machine system (1) for carrying out a method according to one of the preceding claims. [16] Working machine system for carrying out a method according to one of the preceding claims.

Images