Vacuum Pumps and Systems

Abstract

The present invention relates to a vacuum pump (1) comprising: - a number of sensors (11, 13, 15, 17, 19) configured to measure parameters related to the state of the vacuum pump (1), and a control unit (21) configured to acquire measurement values ​​from the plurality of sensors (11, 13, 15, 17, 19), to determine over time and recursively an operating point of the vacuum pump (1) based on the acquired measurement values ​​and an operating mode of the vacuum pump (1) selected by a user from among a plurality of operating modes, and to apply the determined operating point.

Description

[Technical field]

[0001] The present invention relates to the field of vacuum devices configured to place a processing chamber, for example a processing chamber for semiconductor manufacturing, under vacuum. [Background technology]

[0002] Depending on the application, the vacuum equipment, and in particular the vacuum pump, are demanded in different ways. Indeed, some applications require a repeated and fast reduction of the pressure in the vacuum chamber in order to optimize the number of processed pieces, e.g., semiconductors, in a given time course, while other applications may not require such a fast pressure reduction and therefore have fewer constraints on the vacuum pump. Thus, if the settings of the vacuum pump are made at the time of manufacture and are not adjusted according to the user's requirements, the vacuum pump will result in a suboptimal yield for users who require high performance (fast pressure reduction) and a short life or high operating costs for users who do not require such high performance. Furthermore, many parameters of the vacuum equipment may affect the equipment's condition and must be considered to determine the optimal operating point. However, since these parameters are inter-correlated, determining the optimal operating point can be particularly difficult for complex vacuum equipment. Summary of the Invention [Problem to be solved by the invention]

[0003] It is therefore an object of the present invention to provide a solution which allows the settings of the vacuum device to be permanently adapted to provide an operating point according to the user's requirements. [Means for solving the problem]

[0004] That is to say, the invention relates to a vacuum pump comprising: - a plurality of sensors configured to measure parameters related to a condition of the vacuum pump; and - a control unit configured to acquire measurement values ​​from the plurality of sensors, determine over time and recursively an operating point of the vacuum pump based on the acquired measurement values ​​and an operating mode of the vacuum pump selected by a user from among a plurality of operating modes, and apply the determined operating point.

[0005] By determining and adapting the operating point over time and recursively based on measurements and the selected mode, it is possible to optimize the operating point according to user requirements, thus adapting the control of the vacuum pump to a specific application without requiring time-consuming and tedious configuration of the vacuum pump during production to adapt it to customer requirements.

[0006] According to another aspect of the invention, at least some of the parameters related to the condition of the vacuum pump are cross-correlated parameters.

[0007] According to yet another aspect of the invention, the plurality of operating modes refers to the following operating modes: - a mode configured to optimize performance of said vacuum pump; - a mode configured to maximize the useful life of said vacuum pump; - a mode configured to limit a functional cost of said vacuum pump; and - A mode configured to limit energy and / or material consumption.

[0008] According to yet another aspect of the invention, the control unit is configured to apply fuzzy logic to the acquired measurements to determine an operating point for the vacuum pump.

[0009] According to yet another aspect of the invention, the control unit comprises a neural network configured to determine an operating point for the vacuum pump based on the acquired measurements.

[0010] According to yet another aspect of the invention, the parameter related to the state of the vacuum pump is selected from the following: - the temperature of said vacuum pump, - the rotational speed of the rotor of said vacuum pump, - the flow rate of the cooling fluid, - purge gas flow rate, - the pressure level at the input of the vacuum pump or in a chamber connected to the vacuum pump, - Valve settings, - Vibration of parts of the vacuum pump, - the strength of the motor of said vacuum pump; - the noise level of said vacuum pump, and - Ultrasonic waves from the vacuum pump mentioned above.

[0011] According to yet another aspect of the invention, applying the determined operating point includes controlling a supply current to control the rotational speed of the rotor and controlling a flow rate of a cooling fluid.

[0012] According to yet another aspect of the invention, the control unit is configured to take into account current and previously acquired measurements to determine an operating point for the vacuum pump.

[0013] According to yet another aspect of the invention, the control unit is configured to evaluate a remaining useful life of the vacuum pump or a part of the vacuum pump according to the operating point applied over time and to issue a warning signal if the determined remaining useful life falls below a predefined threshold.

[0014] The present invention also relates to a vacuum system comprising: A vacuum chamber; at least one vacuum pump; a plurality of sensors configured to measure parameters related to a state of the vacuum system; a control unit configured to acquire measurements from the plurality of sensors, determine over time and recursively an operating point of the vacuum system based on the acquired measurements and an operating mode of the vacuum system selected by a user from among a plurality of operating modes, and apply the determined operating point.

[0015] According to another aspect of the invention, the plurality of sensors refers to pressure sensors located in various parts of the vacuum system, temperature sensors located in various parts of the vacuum system, a sensor configured to measure the rotational speed of a rotor of the at least one vacuum pump, and a sensor configured to measure the setting of at least one valve.

[0016] According to yet another aspect of the invention, the control unit includes a neural network configured to determine an operating point of the vacuum system based on the acquired measurements.

[0017] According to yet another aspect of the invention, the neural network is configured to generate an adaptation of the operating point over time depending on previous conditions of the vacuum system.

[0018] According to yet another aspect of the invention, the neural network is configured to determine positions and functions of different components of the vacuum system based on past measurements and past states of the vacuum system, and further, the neural network is configured to adapt an operating point of the vacuum system when a new component is introduced to the vacuum system. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a schematic diagram showing a vacuum pump of the present invention. [Diagram 2] FIG. 2 is a schematic diagram illustrating an example of a fuzzy logic modifier definition related to rotor temperature. [Diagram 3] FIG. 3 is a schematic diagram showing a vacuum system of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] The following work is by way of example only. Although this specification refers to one or more embodiments, this is not intended to imply that each reference refers to the same embodiment or that a feature applies only to a single embodiment. Simple features of different embodiments may be combined to provide other embodiments.

[0021] The present invention relates to a vacuum device configured to place a vacuum chamber, typically a processing chamber, under vacuum. By vacuum device, it is meant in particular a vacuum pump or a vacuum system with one or several vacuum pumps and additional components associated with the vacuum pump. In the following, embodiments or features are described based on a vacuum pump or a vacuum system, but the functionality can be applied to any vacuum device.

[0022] Figure 1 is a schematic diagram showing an example of a vacuum pump 1. Vacuum pump 1 can be a primary pump, such as a dry pump, configured such that output 1b is exhausted at atmospheric pressure, as shown in Figure 1. Alternatively, the present invention can be applied to a turbomolecular pump configured to couple with a primary pump or other types of vacuum pumps known to those skilled in the art.

[0023] In the case of FIG. 1, the vacuum pump 1 comprises an input 1a configured to be coupled to a vacuum chamber in which a vacuum has to be set. The vacuum pump 1 comprises a stator 3 on which are arranged two rotors 5 (in FIG. 1 the second rotor is hidden by the first rotor) rotating about two parallel axes X, defining a number of pump stages, in this example five, marked T1, T2, T3, T4, T5. The rotors 5 are arranged to suck in a gas, marked F1, at the input 1a, compress the gas F1 and discharge it towards the output 1b of the vacuum pump 1. The vacuum pump 1 also comprises an electric motor 7 arranged to rotate the rotors 5. The pump stages T1 to T5 are connected in series such that the gas F1 is sent through the different stages T1 to T5.

[0024] The vacuum pump 1 also comprises a purge gas injection device 9, which includes at least one duct configured to be connected to a purge gas supply. The purge gas is denoted F2. In this example, the purge gas injection device 9 has several ducts connected to different stages T1-T5 of the vacuum pump 1. One of the ducts is also connected to the bearing 11. Injecting the purge gas F2 makes it possible to protect various components of the vacuum pump 1 that come into contact with the intake gas F1, which may be aggressive and damage the vacuum pump 1. The vacuum pump 1 may also comprise a fluid circuit (not shown) configured to circulate a cooling fluid in the stator 3, for example adjacent to both rotors 5 or adjacent to the bearing 11.

[0025] The vacuum pump 1 also comprises a number of sensors arranged to measure parameters related to the state of the vacuum pump 1, which parameters may be cross-correlated. The following list of sensors may be used: - temperature sensors 13 which can be arranged in various parts of the vacuum pump 1, adjacent to the different pump stages or adjacent to the bearings 11; - a rotational speed sensor 15 configured to measure the rotational speed of both rotors 5 of the vacuum pump 1; - a flow sensor 17 configured to measure the flow rate of the purge gas; - a flow meter configured to measure a flow rate of a cooling fluid; a pressure sensor 19, which may be arranged in different parts of the vacuum pump, such as the input 1a or the output 1b of the vacuum pump 1, or in a vacuum chamber connected to the input of the vacuum pump 1; - a valve setting sensor configured to send different signals depending on a valve setting, such as a valve configured to control a flow rate of a cooling fluid; - a vibration sensor, such as an accelerometer, arranged on a component such as the stator 3 of the vacuum pump 1; - a magnetic field sensor configured to control an electromagnetic bearing of the vacuum pump 1; - a current sensor, which measures the intensity of the current supplied to the motor of the vacuum pump; - an acoustic sensor to measure the noise level associated with a vacuum pump; - Ultrasonic sensor that measures the ultrasonic waves generated by the vacuum pump.

[0026] The vacuum pump 1 also comprises a control unit 21 configured to acquire measured values ​​from the aforementioned sensors. The control unit 21 is also configured to determine an operating point of the vacuum pump 1 based on the acquired measured values ​​and an operating mode of the vacuum pump 1 selected by a user. The operating modes may be preset during the manufacture of the vacuum pump 1. One of the preset modes is selected by the user depending on the application. The various operating modes can be, for example, any of the following operating modes: a run mode in which the vacuum pump is configured to obtain a target pressure as quickly as possible at the input of the vacuum pump or in a vacuum chamber connected to the input of the vacuum pump; - a life-enhancing mode configured to limit the performance of the vacuum pump when it reaches a potentially damaging level in a parameter, such as a rotor temperature threshold; - a cost limiting mode configured to minimize the functioning cost of the vacuum pump, in which the vacuum pump is controlled to limit energy consumption and consumption of materials such as the amount of purge gas; - Average mode configured to provide higher performance than the enhanced life mode while providing a longer life than the run mode.

[0027] Other modes that may be of interest to the user can also be pre-configured.

[0028] The control unit 21 is also configured to apply the determined operating point, for example by controlling the rotational speed of both rotors 5 via the electric motors 7, setting various valves, controlling the purge gas flow rate and the cooling fluid flow rate.

[0029] Such determination and application of the operating point is performed recursively over time, for example at a predefined frequency, which can be selected depending on the capacities of the different components and can be a few Hz. Such a recursive determination over time makes it possible to adapt the operating point in response to changes in the various measured parameters and to optimize the control of the vacuum pump 1.

[0030] According to a first embodiment, the control unit 21 is configured to apply fuzzy logic commands to determine the operating point.

[0031] The fuzzy logic commands refer to the application of certain rules that depend on the value of the measured parameter with respect to certain thresholds associated with different states of the vacuum pump 1. Furthermore, the different thresholds may refer to ranges of values ​​that may overlap each other, such that the transition between the two states of the vacuum pump 1 is gradual.

[0032] FIG. 2 shows an example of a fuzzy logic rule related to rotor temperature. In FIG. 2, the horizontal axis shows the temperature of the rotor 5, and the vertical axis shows the membership of the set, i.e., the modifiers or variables related to the different temperature ranges. In this example, five modifiers are used to cover the entire temperature range of the rotor 5, but any number of modifiers can be used depending on the required accuracy. The different modifiers refer to a cold state below 30°C, a cold state around 45°C, a medium state around 60°C, a warm state around 75°C, and a hot state above 90°C. As can be seen in FIG. 2, some states overlap with adjacent states.

[0033] Similarly, various parameters determining the state of the vacuum pump 1 can be mapped with modifiers, the number of which can vary for different parameters. Thus, different rule sets can be defined for different operating modes and the rule set associated with the selected mode can be applied to the control of the vacuum pump 1.

[0034] For example, consider a vacuum pump 1 with parameters related to the speed of the rotor 5, the temperature of the rotor 5, and the pressure in a vacuum chamber connected upstream of the vacuum pump 1. The values ​​of these parameters are provided by respective sensors and are mapped to low, nominal, and high variables. In this example, for simplicity, we limit the number of parameters and the number of modifiers for each parameter, but in practice, other parameters of the vacuum pump 1 can also be taken into account, as indicated by the list of multiple sensors above. The first goal of the rule set is to meet or fall below the desired temperature. The second goal is to save energy and reduce wear by reducing the rotor speed when possible. However, if the rotor is cold, a boost mode can be used to improve performance and reduce the time it takes for the vacuum chamber to be fully evacuated.

[0035] A first possible set of rules relating to normal mode is as follows: If the pressure is high and the rotor temperature is low, set the rotor speed high. When the pressure is high and the rotor temperature is high, set the rotor speed to normal. If the pressure is normal, set the rotor speed to normal; If the pressure is low, set the rotor speed low.

[0036] A second possible set of rules related to execution modes is the following: If the pressure is high and the rotor temperature is low, set the rotor speed high. If the pressure is high and the rotor temperature is high, set the rotor speed high. If the pressure is normal, set the rotor speed to normal; If the pressure is low, set the rotor speed low.

[0037] Thus, the operating point is determined by the control unit 21 based on the measurements of the multiple sensors and on the rule set of the selected operating mode. The control unit 21 is then configured to apply the determined operating point. Application of the determined operating point means, for example, control of the supply current, which controls the rotation speed of the rotor 5. Such application may also refer to control of the flow rate of the cooling fluid or to control of other control parameters of the vacuum pump 1. By applying fuzzy logic, the vacuum pump can be efficiently controlled to reach an optimal operating point without requiring significant processing means to take into account the cross-correlations between the various parameters.

[0038] According to another embodiment, the control unit 21 comprises a neural network configured to determine the operating point of the vacuum pump 1 on the basis of the acquired measurements. The neural network is configured to analyze the changes in the state of the vacuum pump 1 and the changes in the measured parameters over time in order to determine the influence of the various control parameters. To set up the neural network, a learning phase may be necessary, including successive steps of activation of the various characteristics of the pump. The permanent analysis achieved by the neural network allows to determine the optimal operating point according to the sought objectives and various constraints. As in the previous embodiment, these constraints refer to the preferred operating mode selected by the user to set various thresholds related to the various parameters, for example the temperature threshold of the operating mode that maximizes the service life is lower than the temperature threshold related to the operating mode that maximizes the performance (minimum time to reach the desired pressure).

[0039] The control unit 21 may be configured to take into account current and past measurements obtained to determine the operating point of the vacuum pump 1. Furthermore, each new measurement may be used to update thresholds or to update the influence of parameters on the state of the vacuum pump 1.

[0040] By analysing the changes in state of the vacuum pump 1 and the effect of various parameters over time, the neural network is able to determine the configuration and function of the various components of the vacuum pump 1 with respect to one another.

[0041] According to a third embodiment, the control unit 21 comprises a combination of fuzzy logic defining the rule set and a neural network configured to adapt values ​​associated with the modifiers defined in the rule set. An operating point is defined based on current and previous measurements and values ​​associated with the modifiers adapted over time. Furthermore, the control unit 21 can automatically adapt the determination of the operating point to replacement of a component of the vacuum pump 1, even if the new component has different characteristics than the replaced component.

[0042] The control unit 21 may be provided using dedicated hardware as well as hardware capable of executing software. If it includes a processor, the control unit 21 may include a single dedicated processor, a single shared processor, or multiple individual processors, some of which may be shared. Furthermore, the explicit use of the term "processor" should not be construed to refer only to hardware capable of executing software, but implicitly includes, but is not limited to, microcontrollers, digital signal processor (DSP) hardware, network processors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), read only memory (ROM) for storing software, random access memory (RAM), and non-volatile storage devices. Other hardware, conventional and / or custom, may also be included.

[0043] Furthermore, the vacuum pump 1 is equipped with communication means, which may be wireless communication means, enabling communication between the various components of the vacuum pump 1, in particular between the control unit 21 and the various sensors, and between the control unit 21 and control means enabling the application of operating points to the vacuum pump 1 (such as controlling the rotational speed of both rotors 5).

[0044] As disclosed above, the present invention can also be applied to a vacuum system including one or several vacuum pumps and possibly additional components such as valves. The various features described for the vacuum pump 1 are also applicable to the vacuum system 1.

[0045] 3 shows an example of a vacuum system 100 including a roughing pump 101, a first turbomolecular pump 103 connected to the roughing pump 101 via a first electronic valve 105, a second turbomolecular pump 107 connected to the first turbomolecular pump 103 via a second electronic valve 109, and a vacuum chamber 115 connected to the second turbomolecular pump 107, in which the three pumps 101, 103, and 107 are connected in series. The vacuum system 100 also includes a first pressure gauge 111 disposed between the roughing pump 101 and the first turbomolecular pump 103, and a second pressure gauge 113 disposed between the first turbomolecular pump 103 and the second turbomolecular pump 107. The vacuum system 100 shown in FIG. 3 is kept simple for the sake of clarity, but in reality such a vacuum system 100 may include many other sensors, such as temperature sensors located in the different vacuum pumps 101, 103, 107, flow sensors associated with the purge gas supplies or cooling fluid circuits of the different vacuum pumps 101, 103, 107, and other types of sensors as disclosed above.

[0046] The vacuum system 100 also comprises a control unit 21 capable of applying fuzzy logic and / or neural networks, as disclosed above for the case of the vacuum pump 1, and configured to recursively determine the operating points of the vacuum system (the drive and control of the various pumps as well as the control of the various valves) based on the various measurements obtained from the various sensors of the vacuum system and according to the operating mode selected by the user.

[0047] In the following example, the control unit 21 is equipped with a neural network.

[0048] When suction of the vacuum chamber 115 begins, the roughing pump 101 is turned on, which causes the pressures on both the first and second pressure gauges 111, 113 to drop.

[0049] Next, the first turbo molecular pump 103 is turned on. As a result, the pressure on the second pressure gauge 113 drops, but the pressure on the first pressure gauge 111 does not drop.

[0050] As a result, the neural network of the control unit 21 can determine that the first pressure gauge 111 is located between the roughing pump 101 and the first turbomolecular pump 103 .

[0051] In the next step of the process, the second turbomolecular pump 107 is turned on, but there is no drop in pressure on either the first pressure gauge 111 or the second pressure gauge 113. The neural network can determine that the second pressure gauge 113 is located between the first turbomolecular pump 103 and the second turbomolecular pump 107.

[0052] Based on this analysis of the different parts of the vacuum system 100, the neural network of the control unit 21 associated with the vacuum system 100 can establish several rules to control the vacuum system 100. For example, the control unit 21 can decide to set the first turbomolecular pump 103 in a standby mode, unless the pressure in the second pressure gauge 113 or the driving force of the second turbomolecular pump 107 increases significantly, in order to reduce wear and power consumption associated with the first turbomolecular pump 103.

[0053] However, the second turbomolecular pump 107 cannot be switched to standby mode, since there is no pressure gauge to monitor the effect of such adaptation in standby mode. Such monitoring of the behavior of the vacuum system 100 can be achieved constantly over time, so that the control unit 21 can adapt to changes during the process. For example, if an increase in the gas load pumped from the vacuum chamber 115 occurs, the standby mode of the first turbomolecular pump 103 can be disabled to ensure the safety of the vacuum system 100.

[0054] Furthermore, in case of changes to the vacuum system 100, for example replacement of an element of the vacuum system 100 due to maintenance or upgrading to a newer element, the control unit 21 can detect such a replacement and adapt the rules and operating point determinations to take into account the characteristics of the new element, so that the vacuum system 100 is always adapted to any changes that may occur during the life of the vacuum system 100.

[0055] Furthermore, in the different embodiments disclosed above, the control unit 21 may be configured to assess the remaining useful life of the vacuum pump 1 or components of the vacuum pump 1 based on the operating points applied over time and / or the measurements acquired over time, and to issue a warning signal when the determined remaining useful life is lower than a predefined threshold. Since in the estimation of the remaining useful life different conditions of the vacuum pump 1 are taken into account over time, this assessment may be very accurate, optimizing the useful life of the components while limiting the risk of failure.

[0056] Such features of the control unit 21 may also be applied in the case of the vacuum system 100 to assess the remaining useful life of components of the vacuum system 100, such as the vacuum pumps 101, 103, 107, the valves 105, 109, or any other components of the vacuum system 100.

[0057] Thus, by monitoring the state of the vacuum pump or vacuum system by means of a plurality of sensors arranged to measure parameters influencing such state, and using these parameter measurements to determine an operating point associated with a selected mode using fuzzy logic and / or neural network processing means, the present invention makes it possible to optimise the functioning of the vacuum pump of a vacuum system according to the application and the requirements of the user. Moreover, such adaptation of the vacuum device to the requirements of the user is achieved automatically by the vacuum device and does not require a tailor-made configuration of the vacuum device at the time of manufacture in order to adapt the configuration of said device to the application and the requirements of the user.

[0058] Furthermore, the use of neural network processing means allows the control of the vacuum system to adapt to any changes in the configuration of the vacuum system. The present invention also provides an accurate estimation of remaining useful life in order to alert the user when an element requires maintenance or replacement before failure occurs. [Explanation of symbols]

[0059] 1. Vacuum pump 1a Input section 1b Output section 3. Stator 5. Rotor 7. Electric motor 9. Purge gas injection device 11. Bearings 13 Temperature sensor 15 Rotational speed sensor 17 Flow Sensor 19. Pressure sensor 21. Control unit 100 Vacuum System 101 Roughing pump 103...First turbomolecular pump 105 First electronic valve 107 Second turbomolecular pump 109...Second electronic valve 111...First pressure gauge 113 Second pressure gauge 115 Vacuum chamber F1...Gas F2 purge gas T1, T2, T3, T4, T5... Pump stages

Claims

1. A vacuum pump (1), characterized in that it comprises: a number of sensors (11, 13, 15, 17, 19) configured to measure parameters related to the state of the vacuum pump (1), and a control unit (21) configured to acquire measurement values ​​from the plurality of sensors (11, 13, 15, 17, 19), to determine over time and recursively an operating point of the vacuum pump (1) based on the acquired measurement values ​​and on an operating mode of the vacuum pump (1) selected by a user from among a plurality of operating modes, and to apply the determined operating point.

2. 2. The vacuum pump (1) according to claim 1, characterized in that at least some of the parameters related to the state of the vacuum pump (1) are cross-correlation parameters.

3. 3. The vacuum pump (1) according to claim 1 or 2, characterized in that the plurality of operating modes means the following operating modes: a mode configured to optimize the performance of said vacuum pump (1); - a mode configured to maximize the service life of said vacuum pump (1), and A mode configured to limit the functional costs of said vacuum pump (1).

4. 3. The vacuum pump (1) according to claim 1 or 2, characterized in that the control unit (21) is configured to apply fuzzy logic to the acquired measured values ​​to determine the operating point of the vacuum pump (1).

5. 3. The vacuum pump (1) according to claim 1 or 2, characterized in that the control unit (21) comprises a neural network configured to determine an operating point of the vacuum pump (1) based on the acquired measured values.

6. 3. The vacuum pump (1) according to claim 1 or 2, characterized in that the parameters related to the state of the vacuum pump (1) are selected from the following: the temperature of the vacuum pump (1), the rotation speed of the rotor (5) of said vacuum pump (1), - the flow rate of the cooling fluid, - flow rate of purge gas (F2), the pressure level at the input (1a) of said vacuum pump (1) or in a chamber connected to said vacuum pump (1), - valve settings, vibration of a part of the vacuum pump (1), the strength of the motor of said vacuum pump (1); the noise level of said vacuum pump (1), and Ultrasonic waves from the vacuum pump (1).

7. 3. A vacuum pump (1) according to claim 1 or 2, characterized in that the application of the determined operating point comprises control of a supply current to control the rotational speed of the rotor (5) and control of a flow rate of a cooling fluid.

8. 3. The vacuum pump (1) according to claim 1 or 2, characterized in that the control unit (21) is configured to take into account current and previously acquired measurement values ​​in order to determine the operating point of the vacuum pump (1).

9. 3. The vacuum pump (1) according to claim 1 or 2, characterized in that the control unit (21) is configured to evaluate the remaining useful life of the vacuum pump (1) or of parts of the vacuum pump (1) according to the operating points applied over time and to issue a warning signal if the determined remaining useful life falls below a predefined threshold value.

10. A vacuum system (100) comprising: A vacuum chamber (115); At least one vacuum pump (101, 103, 107); a plurality of sensors (111, 113) configured to measure parameters related to a state of the vacuum system (100); A control unit (21) configured to acquire measurement values ​​from the plurality of sensors (111, 113), determine an operating point of the vacuum system (100) over time and recursively based on the acquired measurement values ​​and an operating mode of the vacuum system (100) selected by a user from among a plurality of operating modes, and apply the determined operating point.

11. 11. The vacuum system (100) of claim 10, wherein the plurality of sensors (111, 113) refer to pressure sensors (111, 113) located in different parts of the vacuum system (100), temperature sensors located in different parts of the vacuum system (100), a sensor configured to measure the rotational speed of a rotor of the at least one vacuum pump (101, 103, 107), and a sensor configured to measure the setting of at least one valve.

12. 11. The vacuum system (100) of claim 10, characterized in that the control unit (21) includes a neural network configured to determine an operating point of the vacuum system (100) based on the acquired measurement values.

13. 13. The vacuum system (100) of claim 12, wherein the neural network is configured to generate an adaptation of the operating point over time depending on a previous state of the vacuum system (100).

14. 14. The vacuum system (100) of claim 12 or 13, characterized in that the neural network is configured to determine the positions and functions of different components of the vacuum system (100) based on past measurements and past states of the vacuum system (100), and further characterized in that the neural network is configured to adapt an operating point of the vacuum system (100) when new components are introduced into the vacuum system (100).

Images