Control systems for industrial plants

Abstract

We provide control systems for industrial plants. [Solution] The industrial plant (10) has a plurality of industrial devices (11) including sensors and / or data transmission devices and one or more connections to the outside world for communication of data output from sensors and / or data transmission devices, a control processor (4) having a microprocessor and computer-type memory, enabling centralized management of the control system (1) and the industrial plant (10) through an easily intuitive and understandable interface, and configured to convert the data from the industrial devices (11) into human-readable data at runtime, wherein the control processor (4) is connected only to the coordinating device (3) within the control system (1).

Description

Technical Field

[0001] The present invention relates to a control system or device for an industrial plant of the type specified in the preamble of claim 1.

[0002] For example, various control systems for industrial production equipment, including machine tools, processing machines, control machines, and others, are currently known. The term "control" is generally understood to mean the control and / or command of the above industrial plant.

[0003] For example, PLC-type control systems are well known.

[0004] A PLC is a programmable electronic device that executes real-time logic and control functions. To execute these functions, they are equipped with a microprocessor and an input and output system that receives and transmits signals between the industrial machine and the microprocessor and / or electronics that execute the operation.

[0005] In summary, PLCs are versatile and essential industrial automation tools due to their ability to provide precise and reliable process control.

[0006] However, PLCs are more suitable for controlling a single machine rather than an industrial plant.

[0007] Furthermore, they have a generally very complex programming communication interface with the user.

[0008] Next, there is a DCS (Distributed Control System) type control system.

[0009] A DCS control system is a control architecture mainly used in industrial automation to monitor and control processes and plants that include multiple industrial machines and are thus more complex.

[0010] The DCS system uses a bottom-up logic architecture.

[0011] In a DSC system, there are numerous electronic devices that communicate with the individual machines and industrial machinery that make up the plant.

[0012] Such electronic devices transmit data to the operator. The latter can adjust and control the machines connected to the electronic devices through a special interface. The data received and modified by the user are stored in a computer-based server.

[0013] The DSC system also includes an additional layer with a computer processor that communicates with individual electronic devices and can access the stored data. Through this electronic operator, the operator can monitor the entire plant and the proper execution of processes.

[0014] The use of a DCS system improves plant automation and efficiency.

[0015] However, even in that case, the control system is complex for end-users to use and adjust. Furthermore, operators adjust the individual machines that make up the plant without having to worry about their interactions with other machines. In fact, the control of the entire system, which is performed at a higher level, is implemented only downstream.

[0016] Furthermore, there are distributed supervisory control and data acquisition (SCADA) systems, which consist of remote telemetry units (RTUs) that monitor digital and analog field parameters and transmit data to a central monitoring station, and technological solutions used for remote monitoring and control of industrial processes and infrastructure.

[0017] Such a system is also described in patent application US2021 / 0397166 A1.

[0018] The prior art described contains several significant drawbacks.

[0019] In particular, as mentioned above, industrial plant control systems must be implemented in a customized manner specific to each individual plant.

[0020] They are also extremely complex to implement, and to control or repair in the event of a failure.

[0021] Furthermore, industrial plant control systems enable the control and optimization of the entire plant solely from the downstream end.

[0022] In this context, the fundamental technical problem underlying the present invention is to devise a control system capable of substantially overcoming at least a portion of the above-mentioned drawbacks.

[0023] Within the scope of the above technical problems, an important objective of the present invention is to obtain an industrial plant control system that can be easily implemented in the industrial plant itself.

[0024] Another important objective of the present invention is to realize a simple and robust industrial plant control system.

[0025] Another important objective of the present invention is to realize an industrial plant control system capable of optimizing industrial processes.

[0026] The technical challenges and specific objectives are achieved by the industrial plant control system claimed in appended claim 1.

[0027] Preferred embodiments are highlighted in the dependent claims.

[0028] The features and advantages of the present invention will be clarified below by a detailed description of the preferred embodiments of the present invention with reference to the accompanying drawings.

Brief Description of the Drawings

[0029] [Figure 1] A schematic diagram of the operation of the device according to the present invention is shown.

Mode for Carrying Out the Invention

[0030] FIG. 1 shows a schematic diagram of the operation of the device according to the present invention.

[0031] In this document, when measured values, values, shapes, and geometric references (e.g., vertical and parallel) are associated with "approximately" or other similar terms, such as "almost" or "substantially", they are to be understood as excluding errors or inaccuracies in the measured values due to production and / or manufacturing errors, and, in particular, as having a deviation slightly less than that from the associated value, measured value, shape, or geometric reference. For example, when associated with a value, such terms preferably indicate a deviation not exceeding 10% of the value itself.

[0032] Furthermore, when terms such as "first", "second", "upper", "lower", "main", and "sub" are used, they do not necessarily identify an order, priority of relationship, or relative position, but may be used merely to more clearly distinguish different components from one another.

[0033] Unless otherwise specified, as reflected in the following discussion, terms such as "processing", "calculating", "determining", "operating", or the like are to be considered as referring to actions and / or processes of a computer or similar electronic computing device that manipulate and / or transform data represented as a physical, e.g., electronic, quantity in a computer system and / or a record in a memory into other data similarly represented as a physical quantity in a computer system, record, or other information storage, transmission, or display device.

[0034] Unless otherwise indicated, the measurements and data reported in this document should be considered to have been performed in accordance with the International Standard Atmosphere (ICAO) (ISO 2533:1975).

[0035] Referring to the drawings, the control system according to the present invention is shown as a whole by reference numeral 1.

[0036] It is possible to control an industrial plant 10 that includes multiple industrial devices 11.

[0037] The term industrial device 11 means industrial machinery such as machine tools, process machines, control machines, etc., or a part of such machinery, or further, simple industrial tools or devices such as valves, pumps, or others.

[0038] Each industrial device 11 includes general sensors and devices that output data such as operational data, and one or more connections to the outside world for output data communication.

[0039] The control system 1 includes at least one, and preferably more, endpoints 2. Briefly, the endpoints 2 are preferably connected to a coordinating device 3, which is then preferably connected to a control processor 4, which operates remotely (i.e., via a web connection) and / or in the cloud.

[0040] Endpoint 2 is an electronic device equipped with a microprocessor and IT memory that can be read by the microprocessor.

[0041] Each endpoint 2 is also directly connected to at least one industrial device 11 and can be connected to multiple industrial devices 11. The endpoints 2 are then preferably configured to communicate with the industrial devices 11. Such communication may preferably be from endpoint 2 to the industrial device 11 for sending commands, for example, and from the industrial device 11 to endpoint 2 for transmitting operational information and sensor data mounted on the industrial device 11 to endpoint 3. Such connections may be wireless or wired.

[0042] To implement the above connections, each endpoint 2 preferably includes multiple interface connections 20, which are preferably digital, analog, and serial. Preferably, the endpoint 2 also supports different types of industrial sensors, for example, at least one of the following types: PNP, 4-20mA, I2C, and SPI, and preferably all of them.

[0043] Endpoint 2 is responsible for interacting with the industrial device 11, preferably through a different interface: GPIO for digital input and output • SPI and I2C for communication with sensors and other devices ADCs and DACs for reading and generating analog signals • PWM for generating modulated signals.

[0044] Each endpoint 2 also has a first internal connection 21 for interacting with the rest of the control system 1, specifically the coordinating device 3 or alternatively the control processor 4. The first internal connection 21 may be wireless, preferably RF type, or wired, preferably fieldbus type or similar.

[0045] Preferably, each endpoint 2 includes one or more microprocessors or microcontrollers, preferably of ARM type, more preferably ARM-Cortex, configured to connect directly to the peripherals of the industrial device 11. Furthermore, preferably, each endpoint 2 includes integrated firmware.

[0046] Each endpoint 2 preferably analyzes information from the industrial device 11 to which it is connected, either from information contained in a database described later or from an automation function standardization protocol. Preferably, this allows them to autonomously execute read / write commands to peripheral devices according to timing, interrupt, and conditioning logic. Each endpoint 2 is also configured to execute complex control logic, intelligently communicate with the upper layer, and then preferably communicate with the coordinating device 3.

[0047] Each endpoint 2 can also preferably execute and store a variety of commands according to the logic sent by the control processor 4, without the need for any direct reprogramming.

[0048] Each endpoint 2 is also preferably capable of interacting with a variety of industrial devices 11, and more preferably configured to do so. This interaction is facilitated by a database, as described below.

[0049] Each control system 1 also preferably includes a coordinating device 3, as described above.

[0050] The coordinating device 3 is preferably an electronic device having a microprocessor and memory. It is preferably basically an industrial PC and more preferably has a Unix®-based operating system.

[0051] The coordinating device 3 has at least one second internal connection 30 configured to connect to a plurality of first internal connections 21 of a plurality of endpoints 2. The second internal connection 30 is preferably wireless, preferably RF type, or wired, preferably fieldbus type or similar. Preferably, all endpoints 2 of the industrial plant 10 are connected to the same coordinating device 3.

[0052] The coordinating device 3 handles bidirectional communication with endpoint 2 and is connected to the control processor 4.

[0053] The coordinating device 3 is preferably in a master / slave configuration with the endpoint 2. It preferably periodically extracts data from the endpoint 2, checks their status, collects data, sends received commands to them, and processes commands coming from or independently generated by the control processor 4. Furthermore, preferably, the coordinating device 3 acts as a gateway to the control processor 4. The latter is highly stable and accurately ensures the stability of the connection, handshake, and monitoring of the endpoint 2.

[0054] As a result, since the control processor 4 is preferably located remotely and / or in the cloud, the coordinating device 3 plays an active role in locally adjusting the system and monitoring and controlling the system.

[0055] The coordinating device 3 also acts as a bridge between the endpoint 2 and the control processor 4, for example, by converting raw data from the endpoint 2 into structured information and sending it to the control processor 4 for processing and analysis.

[0056] Simultaneously, the coordinating device 3 receives instructions and configurations from the control processor 4, converts them into endpoint-specific commands, and efficiently distributes them.

[0057] Thus, the coordinating device 3 preferably ensures a bidirectional flow of information between the industrial plant 10 and the control processor 4, enabling the control and centralized management of the plant 10 itself in essentially real time or otherwise very fast and timely.

[0058] The coordinating device 3 also preferably forwards commands from the control processor 4 to the appropriate endpoint 2, manages communication with the endpoint 2, monitors the status of the endpoint 2 and communicates it to the control processor 4, and also preferably executes commands specific to its hardware.

[0059] Finally, the coordinating device 3 includes a third data connection with the control processor 4. Since the latter is preferably remote and / or in the cloud, the connection is preferably an internet connection, or, if on-site, a network connection or other connection.

[0060] The control processor 4 comprises at least one microprocessor and an electronic server connected to the microprocessor and containing computer memory that can be read by it.

[0061] Therefore, commands of the control processor 4 can preferably be implemented from any computer connected to the web via an appropriate web address, preferably by security identification, such as a username and password or otherwise.

[0062] Alternatively, direct connection to the control processor 4 is possible, such as via a cable.

[0063] The control processor 4 manages the control system 1 and includes control software that is adjustable and controllable by the user via a web interface using a computer or electronic device, and executes it by command.

[0064] The control software consists of a set of software that enables centralized management and intelligent orchestration of the industrial plant 1 as a whole. In practice, a single server, a single computer, can be connected to multiple coordinating devices 3 even if they are configured in different companies, but each has its own software, or its own users, which is isolated from others, and it is preferable to keep the data of various users isolated and independent.

[0065] The control processor 4, understood as a set of hardware and software, collects, processes, and analyzes data from endpoints 2 through the coordinating device 3, and provides a user interface for plant management, configuration, and monitoring. This interface is easily intuitive and therefore fully understandable and implementable even by non-programmers.

[0066] The control software preferably consists of multiple additional software programs, preferably five different software programs, which are preferably the following: - The first software preferably comprises a design and configuration tool that enables the definition of blueprints for coordinating device 3 and endpoint 2. Such blueprints can address and schedule sensors present and connected to endpoint 2, define instances of endpoint 2 and coordinating device 3 associated with a specific industrial device 11, define a plant network, and do the same. - The second software is preferably a no-code tool, i.e., a tool that can be programmed or configured through variable patterns or choices to define plant logic, generation and configuration of automation routines, and commands sent to endpoints in the field. Such software allows operators to program or configure the behavior of an industrial plant, preferably via an online interface, even without software programming skills. - The third software is preferably a SCADA editor. It is a preferred no-code tool for generating and inserting graphical widgets into a user interface. It then allows the user to fully manage and configure the display of plant data and control widgets. - The fourth software acts as a decoder for plant logic defined by the automation coordination function. Such software addresses and analyzes raw data from endpoint 2 through the coordinating device 3 from the system, using the aforementioned database, or preferably an automation function standardization protocol that enables the analysis of information specific to each sensor or actuator in the field. This analysis process enables accurate and meaningful interpretation of the data, preferably enabling monitoring and control of heterogeneous systems.

[0067] The above database is structured to include information necessary for managing the peripherals of the industrial device 11, and to standardize the information necessary for their operation, even if the industrial device 11 and / or its sensors and data transmission devices differ greatly from each other in both the command modes and the number of commands required for their implementation. Similarly, in the case of sensors in device 11, standardization of the information coming from the sensors allows extremely different types of sensors to be read both as interfaces and in terms of data size and type, and enables the system to automatically convert raw data into human-readable data.

[0068] In detail, endpoint 2 receives raw data from industrial device 11 and transmits the data to control processor 4. The latter retrieves the stored information through the database in the form of code or pseudocode necessary for converting the raw data, and executes this code or pseudocode at runtime to convert the data into a human-readable data type. The term pseudocode means either actual code, or more preferably, a language that is an intermediate between the code and the text list of information.

[0069] Two types of messages are defined through the fourth software, or in any case through the control processor 4: a command sent by the control processor 4 to request an action from endpoint 2, and a response sent by endpoint 2 to notify the result of the command or to send data.

[0070] The commands can be of different types: • Save: Stores the command for later execution. • Task: Execute a saved command at a specific time. • Periodic: Executes saved commands at regular intervals. • Interrupts: Execute saved commands in response to hardware events. • Immediate: Executes the command immediately.

[0071] Furthermore, preferably, the fourth software performs an initial handshake phase for device configuration and synchronization, either through the control processor 4 or otherwise. It also defines a checksum mechanism to ensure data integrity and synchronization between the cloud and the device.

[0072] The above database may also reside in other parts of the software, including the control processor 4, or the coordinating device 3 or endpoint 2. - The fifth software is the core of the management software and is the first software that receives data from the device. It manages a database or protocol described as part of the fourth software on the computer connected to the control processor 4 and the endpoint 2 connected to the industrial device 11. The fourth software also manages message queues with a prioritization system and a retry policy for error handling, and configures messages according to a dedicated protocol. It communicates asynchronously with the coordinating device 3 to ensure a reliable connection. Furthermore, the fourth software stores data from sensors, preferably raw data, monitors the connection to the coordinating device 3, and manages the handshake to establish a reliable connection. This bidirectional mechanism provides significant advantages as it creates various connections between the endpoint 2, the coordinating device 3, and the control processor 4, keeping them in a stable and resilient state.

[0073] The control software also includes a dashboard intended to provide a user interface for plant control defined by third-party software.

[0074] The control system 1 is also configured to implement artificial intelligence algorithms, more specifically machine learning algorithms, and even more specifically neural networks, in order to optimize the operation of the industrial plant 1, as will be described later.

[0075] Ideally, processor 4 allows the user to configure neural network training through the described software and interface, which does not require the application of conventional programming with code, and the neural network training can be configured through a graphical editor or a simple, intuitive selection.

[0076] The artificial intelligence can then rely on data acquired from industrial devices 11, which is obtained from endpoint 2, transmitted to control processor 4 via coordinating device 3, and stored and processed by the latter.

[0077] The network deployment can be carried out at three separate levels, making it possible to apply predictions to the entire control system 1 and industrial plant 10.

[0078] In this way, users can convert their know-how about a particular plant process into deterministic relationships between effective field variables, and then transfer their experience and expertise to a machine learning algorithm, which preferably provides reliable predictions about the performance of the industrial process in question and preferably directly acts on the industrial device 11 to optimize the process.

[0079] The artificial intelligence software that enables the training of the neural network is preferably deployed at three different levels of the control system 1, which will be described later. - The first level of artificial intelligence is configured and operated by the control processor 4. This enables training of the control system 1 with preferably all possible variables, and therefore all control device settings 11, or more preferably user-specified variables. It can also preferably analyze all variables, i.e., all data from the industrial devices 11, taken from endpoint 2 and transmitted to the control processor 4 via the coordinating device 3. Furthermore, the first level of artificial intelligence preferably provides predictive values ​​of process variables useful for decision-making and improving the operation of the industrial plant 10. These predictive values ​​are also provided by the next level. - The second level of artificial intelligence is implemented on the coordinating device 3. Preferably, this level also enables training of the control system 1 and the industrial plant 10 with all variables belonging to the network of the coordinating device 3, i.e., all endpoints 2 and the industrial devices 11 connected to them. This second level of artificial intelligence periodically monitors the data and related field variables coming to it, queries the connected endpoints 2 directly, and acts on possible adjustments that can be implemented in the connected industrial devices 11 by communicating with the appropriate endpoints 2 through model predictions. The second level can also provide predicted values. - The third layer of artificial intelligence is implemented directly on endpoint 2, preferably enabling network training with data from industrial devices 11 directly connected to endpoint 2 itself. In fact, the endpoint preferentially, periodically, and directly monitors data from industrial devices 11 and autonomously acts on possible adjustments that may be implemented in the individual connected industrial devices 11, according to predictions implemented holistically by artificial intelligence. The second layer can also provide predictive values.

[0080] In short, preferably, the control processor 4 can analyze all the information and variables in the control system 1, so it performs most or all of the training of the control system 1, while prediction is performed at all levels, and therefore by both the control processor 4, the endpoint 2, and the coordinating device 3.

[0081] In short, the first level of artificial intelligence has the advantage of having more knowledge and visibility of all data across the entire plant, while the third level of artificial intelligence can act more directly and quickly on individual devices. The second level of intelligence lies somewhere in between the two.

[0082] The control system 1 described above also has an innovative industrial control process in which the industrial plant 10 acts in accordance with the described procedure implemented by the control system 1 described above.

[0083] The control system 1 according to the present invention achieves significant advantages.

[0084] First, control system 1 is simple and effective. It is also less expensive and more robust than known systems such as DCS-type control systems.

[0085] Furthermore, optimizing the industrial plant 10 operated by the control system 1 preferably achieves a reduction in downtime. In fact, artificial intelligence that learns from the operator's experience can predict and prevent process failures or anomalies, thereby reducing unplanned downtime. Research has shown that predictive optimization can reduce downtime by up to 50%.

[0086] The optimization of the industrial plant 10 operated by the control system 1 preferably also achieves an improvement in the efficiency of the plant 10 itself. In fact, artificial intelligence can identify opportunities for improvement in the process, thereby increasing the overall plant efficiency by up to 40 percent.

[0087] The optimization of the industrial plant 10 operated by the control system 1 preferably also achieves quality improvements such as error reduction. In fact, artificial intelligence can detect and correct errors in the process in real time, reducing the number of defective products and production errors by up to 20 percent. Neural networks can perform more accurate and advanced quality checks, ensuring that only high-quality products leave the plant.

[0088] The optimization of the industrial plant 10 operated by the control system 1 preferably also achieves greater flexibility and adaptability of the industrial plant 10 itself. In fact, artificial intelligence can quickly adapt to changes in process or operating conditions, ensuring that the plant remains efficient even in unexpected situations.

[0089] The present invention can be modified to produce different versions that fall within the scope of the inventive concept as defined by the claims.

[0090] In this context, all details may be replaced with equivalent elements, and any material, shape, and dimensions may be used.

Claims

1. A control system for industrial plants, The industrial plant has a plurality of industrial devices, including sensors and / or data transmission devices, and one or more connections to the outside world for communication of data output from the sensors and / or data transmission devices. The control system is A plurality of endpoints, each having a microprocessor and computer-like memory, wherein each endpoint is connected to at least one industrial device among the plurality of industrial devices, communicates with the at least one industrial device, receives the output data, and transmits information and commands to the at least one industrial device, A coordinating device having a microprocessor and computer-type memory, wherein the coordinating device is connected to the plurality of endpoints, A control processor having a microprocessor and computer-type memory, wherein the control processor is configured to enable centralized management of the control system and the industrial plant, for example, through an easily intuitive and understandable human-readable interface, and to convert the data from the industrial device into human-readable data at runtime. Equipped with, A control system in which the control processor is exclusively connected to the coordinating device.

2. The control system according to claim 1, comprising a database containing information necessary for managing the plurality of industrial devices for interacting with the plurality of industrial devices, thereby standardizing the information necessary for the operation of the peripheral devices even if the peripheral devices differ in both the command mode and the number of commands required for the implementation of the peripheral devices.

3. The control system according to claim 2, wherein the database includes instructions for the purpose of standardizing information from the sensor and / or data transmission device portion of the industrial device.

4. The control system according to claim 1, comprising only the components described.

5. The control system according to claim 1, wherein the control processor is configured remotely from the endpoint and the coordinating device, i.e., via a web connection, and the endpoint and the coordinating device are configured in the industrial plant.

6. The control system according to claim 1, wherein each of the endpoints has an ARM-type processor.

7. The control system according to claim 1, wherein each of the endpoints includes PNP, 4-20mA, I2C, and SPI type interface connections for interacting with the industrial device.

8. The control system according to claim 1, wherein the control processor is configured to implement a machine learning algorithm to optimize the operation of the industrial plant.

9. The control system according to claim 8, wherein the coordinating device, the endpoint, and the control processor each implement the machine learning algorithm.

10. The control system according to claim 9, wherein the machine learning algorithm is a neural network.

11. The control system according to claim 10, wherein the control processor is configured to perform at least a large portion of the training related to the machine learning algorithm of the control system.

12. The control system according to claim 11, wherein at least one of the control processor, the coordinating device, and the endpoint is configured to perform predictions via the machine learning algorithm.

13. An industrial plant comprising a plurality of industrial devices including sensors and / or data transmission devices, and one or more connections to the outside world for communication of data output from the sensors and / or data transmission devices, and a control system according to any one of claims 1 to 12.

Images