Method and device for monitoring and / or controlling a machine by means of a digital twin

Abstract

The invention relates to a method and a device for monitoring and / or controlling a machine, in particular a machine in a machine line for handling foods and / or beverages. On an edge device, a code-based model is implemented for simulating physical processes of the machine or a plurality of machines or the machine line. The code-based model is part of a digital twin of the machine. State data of the machine are input into the digital twin of the machine on the edge device and by means of the code-based model of the digital twin a production process of the machine is simulated in real time. The input state data are input parameters for the simulation, and results from the simulation are assigned to the input parameters of the simulation at data level. A temporally static or dynamic state description for the machine is provided based on the input parameters for the simulation and the results from the simulation, and is optionally transferred from the edge device to a server.

Description

[0001] Method and device for monitoring and / or controlling a machine using a digital twin

[0002] The invention relates to a machine and a method and a device for monitoring and / or controlling a machine, in particular a machine in a machine line for processing food and / or beverages. The processing may include filling and packaging the food and / or beverages.

[0003] In modern industry, especially in the bottling and packaging industry, machinery is evolving into increasingly complex systems. With the integration of advanced technologies such as the Internet of Things (IoT), artificial intelligence (AI), and automated control systems, both the performance and complexity of these systems are steadily increasing. This development is leading to a significant increase in the effort required to record, monitor, and describe the dynamic state of a system.

[0004] To ensure a detailed and dynamic status description of as many machines and systems as possible in such a plant, data from a variety of sources must be collected, synchronized, and analyzed. This includes measurements from sensors, operating data, condition monitoring, and many other system parameters. The effort required to collect this data increases with the number of components to be monitored and the frequency of data collection. In highly complex systems, this can quickly lead to a flood of data that can no longer be effectively managed without the use of specialized systems and algorithms.

[0005] A precise and dynamic description of the condition is crucial for several reasons, such as early fault detection, optimization of operations, or extending the service life of individual components or entire machines. Continuous monitoring of the condition of the plant enables anomalies and deviating operating conditions to be detected early, enabling preventive maintenance measures and thus minimizing unplanned downtime. Furthermore, precise knowledge of the plant condition allows for fine-tuning of operating parameters, which can lead to increased efficiency, energy savings, and product quality. Providing a description of the condition in near real-time or with minimal latency is crucial, as it enables immediate response to changes.This is particularly important for critical processes where delays can lead to significant risks or losses.

[0006] Current approaches to data and information provision exist using digital twin technology. The digital twin is a virtual representation of a physical machine, which is continuously updated with data from the real operating environment to simulate, analyze, and predict processes.

[0007] In practice, large amounts of sensor data are continuously collected from the machine system and made available to the digital twin. This requires a robust and reliable infrastructure. Currently, the need to transfer these data volumes to a cloud or central data center leads to various problems, such as bandwidth limitations, latency, security risks, and transmission costs. Time-critical applications suffer from the latency caused by transmission, which can lead to delays in response and decision-making.

[0008] There is therefore a need for a solution that overcomes these disadvantages in the state of the art and thus provides an efficient dynamic state description, especially for filling and packaging machines, lines and plants, based on a digital twin.

[0009] The object is achieved according to the invention by a method according to claim 1 and an edge device according to claim 9. Embodiments and further developments are covered in the subclaims.

[0010] One embodiment of the invention relates to a method for monitoring and / or controlling a machine, in particular a machine in a machine line for processing food and / or beverages. A code-based model for simulating physical processes of the machine and / or a cross-machine code-based model for simulating the physical behavior of a machine line is implemented on an edge device. The code-based model is part of a digital twin of the machine. Machine status data is entered into the machine's digital twin on the edge device, and a production process of the machine is simulated in real time using the code-based model of the digital twin. The entered status data are input parameters for the simulation, and results from the simulation are assigned to the simulation input parameters at the data level.A temporally static or dynamic state description for the machine is provided based on the input parameters for the simulation and the results from the simulation and can optionally be transferred from the edge device to a server.

[0011] A further embodiment of the invention relates to a computer device for monitoring and / or controlling the machine.

[0012] Exemplary aspects of the invention are illustrated in the drawings. They show:

[0013] Figure 1 : a diagram showing an overview of the essential elements and the basic structure of the invention;

[0014] Figure 2: an exemplary flow chart for a method for operator-guided commissioning of the machine;

[0015] Figure 3: an exemplary system configuration for PET containers and

[0016] adhesive packaging;

[0017] Figure 4: an exemplary system configuration for PET containers and

[0018] shrink packer;

[0019] Figure 5: an exemplary system configuration for cans or glass bottles; and

[0020] Figure 6: an example system configuration for cans.

[0021] The present invention aims to provide a temporally static or dynamic state description for filling and packaging machines, lines, and systems based on a digital twin. Typically, three or four levels are mentioned in the literature, where data acquisition, data processing, analysis and diagnostics, fault prediction, and intelligent decision-making take place. These levels include, for example, the machine controller, an edge device, and a server.

[0022] Digital twins of machines are generally virtual replicas of physical assets that are fed with data in real time. This allows for the precise monitoring and analysis of a machine's condition and the making of predictions about future operation. Data is generally collected by sensors attached to machinery. These sensors can record a wide range of information, such as temperature, pressure, vibration, humidity, and much more.

[0023] Based on the data, a digital twin can implement machine learning and / or artificial intelligence algorithms to draw conclusions from the analysis of machine data. This allows potential problems to be diagnosed and future failures to be predicted. Predictive maintenance, for example, can also predict when a part might fail and proactively replace it before a downtime occurs.

[0024] As mentioned in the discussion of the state of the art, digital twins, due to their use of machine learning algorithms, require very large amounts of data to create a similarly detailed picture of the processes running on the machines and conveyors. These disadvantages can be overcome with the concepts described here.

[0025] Figure 1 shows a diagram providing an overview of the essential elements and basic structure of the invention. A machine line 100, for example, for bottling beverages, comprises one or more machines 101, 102, and 103. The machines 101-103 can be functionally connected to one another, for example, via conveyor belts 110 and 112, which ensure a transport flow of materials. However, the present invention is not limited to a machine line 100 or to a plurality of interconnected machines 101-103. Concepts of the invention can also be applied to a single machine.

[0026] The machine line 100 is connected to an edge device 120 and, via the edge device 120, to a cloud or a server 130. According to embodiments, the edge device 120 may be physically integrated into the server 130 or represented by a virtual instance on the server 130.

[0027] Edge device 120 consists of one or more computers that perform data processing tasks at or near the source of data generation, i.e., directly at the edge of the network. They are an important component for bringing computing power closer to where it's needed, such as to machines 101-103 and / or conveyor belts 110, 112, i.e., closer to the machine line 100 that generates the data.

[0028] The edge device 120 can typically include several core components essential to the edge device's tasks, such as processors, memory, network components, interfaces, an operating system, and software. The hardware, and in particular the processor(s) in the edge device 120, are designed, for example, to provide high computing power so that computational operations on the machine data can take place in real time. These processors can range from conventional CPUs to specialized microcontrollers to advanced GPUs, depending on the requirements of the specific application.

[0029] According to embodiments of the invention, a digital twin of the machine system 100 or a part of the machine system 100 is implemented at least in part on the edge device 120. This means that at least a part of a digital twin of the system 100 or one of the machines 101-103 or the conveyor belts 110, 112 or (sub)components of the system 100 can be implemented on the edge device 120. Another part of the digital twin can be implemented, for example, on the server 130.

[0030] The following description uses the term “digital twin of a machine”, but “machine” refers to either one or more of the machines 101-103, or the conveyor belts 110, 112 or (sub-)components of the system 100.

[0031] The part of the digital twin implemented on the edge device 120 includes a code-based model for simulating physical processes of the machine.

[0032] This physical model can be capable of mapping a wide variety of sections in the container transport process or accurately simulating a wide variety of machine functions in the system 100. A key advantage of the physical models is that they are code-generated. The code can be run on the edge device 120 (i.e., on a machine-level computing unit).

[0033] The physical model can obtain information about the products currently being filled / produced from a line management system (e.g., from server 130). Additional optional data sources can include live video (where possible) to obtain information and the status on a section of the line, or other measurement signals transmitted to the digital twin via industrial communication methods. This information can be processed in the simulation model, and the results, depending on the input data, can be transmitted via an interface to cloud 130. This offers the advantage that only a manageable portion of the data needs to be transferred to cloud 130 to provide monitoring and / or control of a machine.The models located in the digital twin can, for example, simulate the production process dynamically and in real time by continuously or at specific events or intervals collecting data from the controls of the machines, the drives or their sensors, the installed sensors such as pressure, position, temperature for the machine and / or the environmental properties of the machine as well as image and video recordings for the machines and transport equipment and / or the operator movements / interventions.

[0034] According to embodiments, this data can be collected by the edge device 120 or by a computer system located near the production line (data collection and processing system). According to embodiments, this data can also incorporate information from a central line management and / or production planning system in order to obtain a direct correlation between the machine and conveyor states and the currently filled and / or packaged product.

[0035] According to some embodiments, the data collected in this way can be preprocessed and prepared so that this data can be used as input parameters for the parallel physical simulations. This can also be done on the edge device, as shown in Figure 1.

[0036] The results from the simulations can, for example, be clearly assigned to the input parameters of the simulation models at the data level and transferred to the next level of the digital twin. The models used in the simulations can thus function as virtual sensors or soft sensors capable of describing conditions in the manufacturing process or in process engineering processes that would not be representable in measurement technology without the simulations or would only be possible with considerable additional effort. Examples of this could be the temperature inside a component, the temperature distribution in a tank filled with fluid, and / or the stress and damage and reduction of a component's service life due to improper use.

[0037] This has the advantage that only a limited amount of data needs to be transferred from the effective level of the edge device 120 (level 2) to the effective level on the server 130 (level 3), but the data can always be used to represent a comprehensive picture of cause and effect from machine data and simulation results. For example, the digital twin can make it possible to simulate the occupancy level of the transport equipment from the current status data of the machines and / or the speeds of the conveyor belts, without these being visible to a camera. According to another example, the heating process of a preform (i.e., a PET bottle blank) can be represented from the machine output and the recipe of a heating furnace, for example for different material parameters.In contrast to the information from images or sensors, the behavior inside the object under investigation can be represented using the simulation results. For example, it is possible to map the temperature gradient across the wall thickness of a preform or the heating of the container contents in the pasteurizer based on the machine data and the physical model that simulates the heating in parallel.

[0038] According to one embodiment, the simulations on the physical models can run significantly faster than the process takes in reality. This makes it possible to operate the digital twin as a feedforward control or to develop and operate control algorithms based on the digital twin. The data connection from the data sources (i.e., from the system 100) to the edge device 120 can ideally be designed such that data, information, and / or software code, software artifacts, or parameters can be exchanged in both directions, simultaneously or sequentially. The edge device 120 can be configured with regard to the number of processors, clock speed, RAM, GPU, and storage to act as a computing machine, making the raw data or previously processed data available to the models, and to execute the simulations in a simulation area (separate) from the data preparation area.

[0039] According to embodiments, the physical models can be located in so-called containers, which contain the necessary software infrastructure for communication, execution, evaluation, and provision of simulation results with the models. In alternative embodiments, however, it is also conceivable for several or all models to be located in one container. According to alternative embodiments, the models can be provided as an FMU ("Functional Mockup Unit") and executed on the edge device 120 with suitable software. It is also possible for containers / FMUs to communicate with other provided containers / FMUs.

[0040] In the example described here, 120 different types of models are located on the edge device. This allows for the use of data-based models, logic models, and even physical models, which enable physical simulations to be performed in real time based on the provided data. Real time in the sense of control engineering can be considered to be in the range of 1-5 ms. However, it may also be sufficient to define real time in the sense of the actual process time (ms, sec, min). Depending on the application, the models located on the edge device 120 are always fast enough to determine the necessary information in the simulations and to report the resulting measures back to the real system or the higher-level layer.

[0041] According to preferred embodiments, the models implemented on the edge device 120 can be used for active control of the connected real systems 100. In some cases, this can lead to the models located on the edge device 120 being used as part of a model predictive (MPC) approach. The simulation based on the model used can run significantly faster than the real process in the system 100.

[0042] The digital twin can thus look somewhat into the future, predicting the ideal next step and reporting it back to the real system 100. In addition to its use as a model predictive controller, where the time horizon is usually on the order of several milliseconds, the approach described can also be used to represent longer periods in the future. This is where physical models come into play. Unlike purely data-based models, so-called predictors and / or regression models, they do not require a complex data pool to generate for a prediction, but can represent the new situation with the appropriate, new boundary conditions via a simulation.

[0043] These simulation results can be combined with statistical data regarding occurring disruptions, which can provide a very detailed look into the future, on the basis of which, for example, product change scenarios can be planned, operators can be put on standby, and raw materials can be provided.

[0044] Embodiments can also be used for condition monitoring and predictive maintenance and / or self-healing, since the use of physical models allows the impact of a malfunction on the process to be directly assessed. For example, if a heating element in a preform heating furnace fails, the effect on the preform heating process can be immediately determined based on the physical simulation. The wiring of the remaining elements can be optimized using the model and transferred back to the real control system.In the area of ​​predictive maintenance, the use of physical models eliminates the need for complex fault data generation. Instead, simulation results can be compared during operation with a database or a correlation model that can establish a direct relationship between the load in the current cycle and the maximum tolerable load cycles. For example, for a particularly stressed part, the stress in MPa can be calculated. This stress results from the load, the acceleration from the drives, and the mass of the product being transported. An inline comparison with the data from the Wöhler diagram for the corresponding material allows the remaining service life to be determined very precisely.

[0045] According to embodiments, level 2 (edge ​​device 120) and level 3 (server / cloud 130) are expanded to include the ability to perform simulations with physical models based on the available data. This makes it possible to generate a significantly more detailed view of events in and on the machine line 100 with the same amount of data. The potential ranges from the model-based determination of occupancy levels to the simulation of the heating behavior of filled products.

[0046] Figure 2 shows an exemplary flowchart of a method 200 for monitoring and / or controlling a machine. The method starts at step S202 with the implementation, on the edge device 120, of a code-based model for simulating physical processes of the machine. As already described, the code-based model is part of a digital twin of the machine. The code-generated model can be created and parameterized, for example, in an automated, semi-automated, or manual workflow in the cloud 130 and transferred to the edge device 120 via a data line.

[0047] In step 204, machine status data is collected by the machine's digital twin on the edge device 120 and / or made available to the digital twin. The status data may include sensor data, but may also include other data sources, such as video streams, as described above.

[0048] In an optional step S206, the state data can be preprocessed. Preprocessing the state data can include filtering sensor data based on relevance to a required output of the code-based model. Additionally or alternatively, the preprocessing can also include discarding incomplete data sets. In step S208, a production process of the machine is simulated in real time using the code-based model of the digital twin. The input state data are input parameters for the simulation, and results from the simulation are assigned to the input parameters of the simulation at the data level. In step S210, a temporally static or dynamic state description for the machine is then provided based on the input parameters for the simulation and the results from the simulation and (optionally) transmitted from the edge device 120 to the server 130 in step S212.

[0049] In a further step, the state description can be processed and used as an input variable for a control and / or for closed-loop control on the real PLC and / or an industrial PC.

[0050] The simulation of the physical processes on the edge device 120 can run faster than the actual process in the machine. In this case, in step S214, the digital twin on the edge device 120 can be used as a feedforward control for the machine. The edge device 120 can generate control signals based on the temporally static or dynamic state description and send them to a control device of the machine.

[0051] The following Figures 3 to 6 describe various exemplary system configurations for different bottling plants in which the invention, or at least parts and aspects of the invention, can be implemented. The description of Figures 3 to 6 is intended only to provide a general overview of machines for which status data can be collected, based on which the LLM can process user requests.

[0052] Figure 3 shows an exemplary system configuration 1000 for PET bottles or PET containers and adhesive packs. As shown in Figure 3, the system configuration 1000 comprises various modules that form a line, at the end of which finished PET containers are dispensed in the form of a pack on pallets. Some of the modules and machines may be optional, and the invention is not limited to the exact form and arrangement of the system configurations.

[0053] The system configuration 1000 comprises an oven 1002 for preforms, a preform sorter with a feeding machine 1004, and a blow molding machine 1008. The modules 1002, 1004, and 1008 generally form a stretch blow molding machine in which PET containers are produced and formed from a starting material. The produced PET containers are forwarded to a filler 1010, where the bottles are filled. The filler can optionally include a rinser. Various particles such as dust, cardboard, or remnants of wooden pallets can settle in the preforms during storage or transport. These can be removed with the rinser. A closer can be arranged at the end of the filler, by means of which the PET containers are closed after filling.

[0054] Optionally, the system configuration 1000 can include a rotating device downstream of the filler 1010, which is used for hot filling of the PET containers. Via one or more conveyor belts 1016, which can also include a buffer 1018 for intermediate loading of filled containers, the filled PET containers are conveyed to a separator 1020 and then to a drying device 1024, in which the PET containers are dried.

[0055] After drying, the PET containers are conveyed to a labeling machine 1026. The labeling machine 1026 can be designed for various labeling techniques, such as labeling using hot melt, cold melt, self-adhesive labels, or sleeves. After the PET containers have been printed or labeled, they are conveyed through a second drying device 1028, a line distributor 1030, conveyor belts 1032, an adhesive pack production line 1034, and a curing section to a handle applicator. In the adhesive pack production line 1034, the PET containers are grouped into specific group sizes and packaged into a pack, such as a "six-pack." In the handle applicator, a carrying handle is attached to the pack, which allows for comfortable carrying of the pack.The finished containers are then arranged accordingly by a robot 1042 for layer production and packed on pallets by a palletizer 1044.

[0056] In system configuration 1000, so-called format trolleys or format racks can be arranged on various modules and machines to provide quickly interchangeable format sets for short changeover times and automatic tool changes. Examples of format trolleys are format trolley 1006 for blow molding machine 1008, format trolley 1012 for filler 1010, format trolley 1022 for labeling machine 1026, format trolley 1038 for adhesive pack production 1034, and format trolley 1046 for palletizer 1044.

[0057] Figure 4 shows another example system configuration 1100 for PET containers and shrink packers. System 1100 in Figure 4 includes many of the modules and machines from system configuration 1000 in Figure 3, but there are some differences. Therefore, the description of the modules already described in connection with Figure 3 is omitted for Figure 4.

[0058] A key difference between the two exemplary system configurations 1000 and 1100 is that the labeling machine 1126 with the labeling modules 1127 can be installed downstream of the blow molding machine 1008 and upstream of the filler 1008. For this purpose, the system configuration 1100 can comprise six transport lanes 1150 into which the PET containers can be pushed. After the PET containers have pushed into one of the six lanes 1150, they are conveyed into the film wrapping module 1152 and then into the shrink tunnel 1154.

[0059] Figure 5 shows an example system configuration 1200 for cans or glass bottles. The example system configuration 1200 from Figure 5 again has some similarities to the system configurations 1000 and 1100 from Figures 3 and 4, and the description of the system configuration is therefore limited to the differences between the system configurations.

[0060] As shown in Figure 5, the exemplary system configuration can include two separate feeders. A first feeder, on the left in Figure 5, shows a branch for cans or, optionally, a partial branch for new, reusable bottles. The containers, i.e., cans or new bottles, are fed into the machine by a depalletizer 1302, where they are guided via conveyor belts to the filler 1010. A second feeder, on the right in Figure 5, shows a partial branch for reusable bottles, which are fed into the system by a reusable sorting system (not shown).

[0061] In the case that the already used reusable bottles are introduced into the system 1200 via the sub-branch for reusable bottles, the reusable bottles first pass through the cleaning machine or washing machine 1304. Another possible difference in the exemplary system configuration 1200 is the transfer packer 1306 after the labeling machine 1026. The transfer packer can sort the bottles or cans into a carton clip application or into crates, or both.

[0062] Figure 6 shows an exemplary system configuration 1300 for cans, in which the elements already described in the other system configurations are no longer described. The cans in system configuration 1300 are fed into the depalletizer 1302 from a magazine 1402 containing cans. After passing through the filler and being filled, the cans are closed by means of a closure magazine 1404 and transported further along the system 1400 via the conveyor belts, as described above. The optional pasteurizer 1408 can be bypassed via the bypass 1412 if it is not required. In the pasteurizer 1408, the freshly filled products can be pasteurized for preservation.

[0063] In contrast to plant configurations 1000, 1100, and 1200, the exemplary plant configuration 1300 shows various tanks for corresponding consumables, such as tanks 1410 with rinsing liquid and / or the filling product and tanks 1406 with belt lubricant. These tanks can also be included in the exemplary plant configurations described above. For example, the chemical products 106 that are fed from the mixer 110 to the machines can be stored in tanks 1406 and 1410.

Claims

CLAIMS 1. A method for monitoring and / or controlling a machine (101-103), in particular a machine in a machine line (100) for treating food and / or beverages, the method comprising: Implementing (S202), on an edge device (120), a code-based model for simulating physical processes of the machine or of multiple machines in the machine line or of sections of the machine line or of the entire machine line, wherein the code-based model is part of a digital twin of the machine; Providing (S204) state data of the machine for the digital twin of the machine on the edge device; Simulating (S208), using the code-based model of the digital twin, a production process of the machine in real time, whereby the input state data are input parameters for the simulation and results from the simulation are assigned at the data level to the input parameters of the simulation; and Providing (S210) a temporally static or dynamic state description for the machine based on the input parameters for the simulation and the results from the simulation.

2. The method of claim 1, wherein the simulation of the physical processes on the edge device (120) runs faster than the real process, and wherein the method further comprises: Operating (S214) the digital twin on the edge device as a feedforward control for the machine or for several machines or for sections of the machine line or for the entire machine line, wherein the edge device generates control signals based on the temporally static or dynamic state description and sends them to a control device of the machine.

3. The method according to claim 1 or 2, wherein the code-based model is implemented in a container or as an FMU on the edge device (120), which comprises necessary software infrastructure used for communication, execution, evaluation and / or provision of the results from the simulation with the models; and / or wherein the code-generated model is created and parameterized in an automated or semi-automated or manual workflow on a server (130) and is transferred to the edge device via a data line.

4. The method according to any one of claims 1 to 3, wherein the simulation by means of the code-based model is further a basis for a model-predictive control of the machine, and / or wherein the results of the simulation are further combined with statistical data relating to occurring disturbances in order to predict a future state of the machine and / or the results of the simulation are incorporated into a process control as virtual sensor signals.

5. Method according to one of claims 1 to 4, wherein the temporally static or dynamic condition description forms the basis for condition monitoring of the machine and / or predictive maintenance of the machine.

6. A method according to any one of claims 1 to 5, further comprising: Preprocessing (S206) the state data; and / or Transmitting (S212) the temporally static or dynamic state description from the edge device to a server 130; wherein the preprocessing of the state data comprises: Filtering sensor data based on relevance to a required output of the code-based model, and / or Discarding incomplete data sets, and / or a stationarity analysis, and / or Determination of static parameters.

7. Method according to one of claims 1 to 6, wherein the status data comprises data from the controls of the machines, data from drives or their frequency converters, data from installed sensors, in particular sensors for pressure, position, temperature for the machine, data of the environmental properties of the machine, data from image and / or video recordings for the machines and / or transport devices and / or from operator movements and interventions, and / or data from a central management and / or production planning system.

8. The method according to any one of claims 1 to 7, wherein the edge device (120) is a machine-level computing unit and further comprises a data connection from the data sources to the edge device, which is designed such that data, information and / or software code or software artifacts or parameters can be transmitted in both directions simultaneously or sequentially.

9. A computer device (120) for monitoring and / or controlling a machine, in particular a machine in a machine line for treating food and / or beverages, the computer device comprising: a memory for storing computer code; a network interface for receiving and transmitting data; and a processor for executing computer code, the computer device being designed to: Implementing (S202), in memory, a code-based model for simulating physical processes of the machine or of several machines in the machine line or of sections of the machine line or of the entire machine line, wherein the code-based model is part of a digital twin of the machine; Receiving (S204), via the network interface, status data of the machine and entering the status data into the digital twin of the machine; Simulating (S208), using the code-based model of the digital twin, a production process of the machine in real time, whereby the input state data are input parameters for the simulation and results from the simulation are assigned at the data level to the input parameters of the simulation; and Providing (S210) a temporally static or dynamic state description for the machine based on the input parameters for the simulation and the results from the simulation.

10. The computer device (120) of claim 9, wherein the simulation of the physical processes runs faster than the real process, and wherein the computer device of the further process is adapted to: Operating (S214) the digital twin as a feedforward control for the machine, wherein the computer device based on the temporally static or dynamic State description generates control signals and sends them to a control device of the machine; or Operating (S214) the digital twin as a virtual sensor for the machine and / or for a process running on the machine, whereby the simulations running in the digital twin provide data for a respective process or state that could not otherwise be generated or could only be generated through the use of complex and expensive measurement technology.

Images