Digitized reporting of technical systems

Abstract

The present disclosure relates at least to digitized reporting of a technical system. Methods, computer program products and computer systems for converting controller data uploaded to a cloud into digital reports stored in the cloud and accessible via an online portal are disclosed. One or more data packets are received from a controller of the compressor over a network, the one or more data packets including data associated with technical parameters of the compressor in a given state of the compressor. And converting the data contained in the data packet into a digital report of the given state of the compressor in a human readable format. The digital reports are stored in one or more databases accessible through an online portal associated with the computer system. Using one or more artificial intelligence models and machine learning techniques, digital reports are used as input datasets to improve fault event prediction and corrective measure response times for technical systems.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of and priority to U.S. Provisional Application No. 63 / 508,490, filed on June 15, 2024, entitled “Digitized Reports of Technical Systems.” Technical Field

[0003] The following relates to digital reporting of technical systems, and more specifically, to converting controller data uploaded to the cloud into digital reports stored in the cloud and accessible via an online portal. In addition, digital reporting utilizes artificial intelligence to improve technical systems. Background Art

[0004] Where industrial equipment or safety-critical equipment (such as medical gas compressors used in hospitals) is used, most jurisdictions require regular monitoring of the equipment, and all jurisdictions at least recommend it. Summary of the invention

[0005] Embodiments of the present invention provide a method, computer program product, and computer system for converting controller data uploaded to the cloud into digital reports stored in the cloud and accessible via an online portal.

[0006] One or more data packets are received from a controller of the compressor over a network, the one or more data packets including data associated with a technical parameter of the compressor at a given state of the compressor. The data contained in the data packets is converted into a digital report in a human-readable format for the given state of the compressor. The digital report is stored in one or more databases accessible through an online portal associated with the computer system. One or more artificial intelligence models and machine learning techniques may be utilized to improve the technical system using the digital report as an input data set, such as improving fault event prediction and corrective action response time of the technical system. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A computing environment according to an embodiment of the present invention is depicted, which includes an example of an environment for executing at least some of the computer code involved in performing the methods of the present invention.

[0008] Figure 2 According to an embodiment of the present invention Figure 1 A block diagram of the modules included in the code contained in the system.

[0009] Figure 3 is a flow chart of an embodiment of a method for converting controller data into digital reports according to an embodiment of the present invention.

[0010] Figure 4 is a flowchart of a process for predicting a fault event and taking corrective action according to an embodiment of the present invention, wherein the operations of the flowchart are Figure 2 One or more modules in the .

[0011] Figure 5 is a flowchart of a process for simulating changes to a compressor system based on the output of a trained AI model according to an embodiment of the present invention, wherein the operations of the flowchart are Figure 2 One or more modules in the .

[0012] Figure 6 is a flowchart of a process of controlling a compressor system based on the output of a trained AI model according to an embodiment of the present invention, wherein the operations of the flowchart are Figure 2 One or more modules in the . DETAILED DESCRIPTION

[0013] Overview

[0014] Where industrial equipment or safety-critical equipment (such as medical gas compressors used in hospitals) is used, most jurisdictions require regular monitoring of the equipment, or at least recommend it in all jurisdictions. To meet the monitoring requirements, organizations using the equipment hire trained technicians to visit the equipment daily, check alarms, record sensor reading levels, and perform general inspections based on equipment-specific checklists. Currently, there is no unified standardized inspection and recording method, and organizations develop their own methods to inspect equipment and record inspections. For example, a clipboard can be attached to each piece of equipment, where the technician can sign at certain locations to indicate that the equipment has been inspected at a given time / date. Another example is a single inspection checklist covering all equipment that is manually signed. Sometimes, organizations are audited to provide evidence of compliance with monitoring requirements. It is often time-consuming to locate complete inspection records, and sometimes they cannot be found due to the method used to record the inspection. In addition, current inspection methods ignore various technical parameters required to accurately assess the operating condition of the equipment in the current or foreseeable future.

[0015] Embodiments of the present invention standardize and improve conventional inspection and recording methods using an online portal that is accessible to members of an organization that uses industrial equipment or safety-critical equipment that requires regular monitoring. Data packets containing data associated with technical parameters are periodically uploaded from the controller / computer system of the equipment directly to one or more servers in the cloud over a network, which convert the data in the data packets into digital reports of the equipment status / condition. To trigger the upload of the data packets, a technician physically interacts with a graphical user interface (GUI) to log into an online portal associated with one or more servers and navigates the GUI to automatically upload the data packets to the cloud. The requirement for physical interaction with the GUI ensures that the technician is present to initiate the upload. Data packets received by one or more servers in the cloud are converted from key-value pairs into digital reports in human-readable form. The digital report includes the timestamp of the upload and various sensor readings, alarm conditions, unit operating hours, the technician's name, the location of the equipment, etc. Organization members can access the data via an online platform such as ) access digital reports stored on cloud servers from local machines, and the online platform operates as a central digital location for managing medical assets, inventory, and teams.

[0016] In addition, embodiments of the present invention improve machine learning algorithms for predicting equipment failure events, simulating corrective actions, and determining corrective actions to be implemented on equipment. The trained artificial intelligence (AI) model is used to output predicted failure events of the equipment and corrective actions taken to avoid failure events; data from data packets automatically uploaded from the equipment controller and / or data from digital reports are used as inputs to the trained AI model. The trained AI model is trained by feeding data sets related to actual failure events of the equipment and known solutions to correct or avoid failure events. The output of the trained AI model is used to repair, modify, maintain, change, etc. the equipment before a failure event occurs or worsens.

[0017] In addition, the output of the trained AI model is used to enhance or otherwise change the GUI of the equipment controller to display a new icon indicating a potential fault event. The change in the GUI alerts the technician to the potential problem, and physical interaction with the new icon will initiate a recommended solution to the problem, which can be downloaded to the equipment controller or other computer from one or more servers in the cloud that utilize the AI ​​model trained for the solution. To ensure that the detected potential fault event is not overlooked or ignored, new data packet transmissions from the equipment controller can be blocked (e.g., at the next interval) until confirmation is received from the equipment controller that the physical input associated with the new icon of the graphical user interface has been received; a notification is immediately sent to the online portal that the transmission of the new data packet is blocked. Alternatively, a new data packet transmission of the new state of the compressor can be accepted and a digital report created, but the equipment controller can display on the graphical user interface that the transmission is blocked.

[0018] Additionally, the output of the trained AI model is fed into a digital computer model of the compressor to simulate adjustments to compressor / medical air system parameters before adjustments are implemented on active compressors located in the facility.

[0019] In a further embodiment, the output of the trained AI model is used to automatically control active compressors located at the facility by implanting the recommended changes.

[0020] Computing Environment

[0021] Figure 1 A computing environment according to an embodiment of the present invention is depicted, which includes an example of an environment for executing at least some of the computer code involved in performing the method of the present invention. The computing environment 100 includes an example of an environment for executing at least some of the computer code involved in performing the method of the present invention, such as code 200 for converting controller data into digital reports. The computing environment 100 includes, for example, a computer 120, a network 140 (such as a wide area network (WAN)), a device controller 180, and a remote server 170 with a database 171. The computer 120 includes a processor 130, a memory 131, a storage device 141 including an operating system 150 and the code 200. The computer 120 is connected to the remote server 170, the device controller 180, and a digital model of the technical system 145 via the network 140. The digital model of the technical system 145 can be located in the storage device 141 alternatively or additionally.

[0022] Cloud computing is a service delivery model for convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be quickly provisioned and released with minimal management cost or interaction with the service provider. This cloud model can include at least five characteristics, at least three service models, and at least four deployment models.

[0023] Features are as follows:

[0024] On-demand self-service: Cloud consumers can unilaterally and automatically provision computing capabilities such as server time and network storage as needed without human interaction with the service provider.

[0025] Broad network access: Capabilities are available and accessible over the network through standard mechanisms that facilitate the use of heterogeneous thin-client or thick-client platforms (e.g., mobile phones, laptops, and PDAs).

[0026] Resource pooling: The computing resources of a provider are pooled to provide services to multiple consumers through a multi-tenant model, where different physical and virtual resources are dynamically allocated and reallocated based on demand. This is location-independent because consumers generally do not control or know the exact location of the provided resources, but may be able to specify the location at a higher level of abstraction (e.g., country, state, or data center).

[0027] Rapid elasticity: Capacity can be provisioned quickly and elastically, and in some cases automatically, to scale up quickly and release quickly to scale down quickly. To consumers, the capacity available for provisioning often appears to be unlimited and can be purchased in any quantity at any time.

[0028] Measurable services: Cloud systems automatically control and optimize resource usage by leveraging metering capabilities at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency to both providers and consumers of the services being used.

[0029] The service model is as follows:

[0030] Software as a Service (SaaS): The capability provided to the consumer is to use the provider's applications running on the cloud infrastructure. These applications can be accessed from a variety of client devices through a thin client interface such as a web browser (e.g., web-based email). The consumer does not manage or control the underlying cloud infrastructure, including the network, servers, operating systems, storage devices, or even individual application capabilities, with the possible exception of limited user-specific application configuration settings.

[0031] Platform as a Service (PaaS): The capability provided to consumers is to deploy applications created or acquired by consumers using programming languages ​​and tools supported by the provider onto the cloud infrastructure. Consumers do not manage or control the underlying cloud infrastructure (including networks, servers, operating systems, or storage devices), but control the deployed applications and possible configuration of the application hosting environment.

[0032] Infrastructure as a Service (IaaS): The capabilities provided to consumers are the provisioning of processing, storage, networking, and other basic computing resources, where consumers can deploy and run arbitrary software, which can include operating systems and applications. Consumers do not manage or control the underlying cloud infrastructure, but do control the operating system, storage, deployed applications, and may have limited control over select network components (e.g., host firewalls).

[0033] The deployment model is as follows:

[0034] Private cloud: Cloud infrastructure is operated by an organization only. It may be managed by the organization or a third party and may exist on-premises or off-premises.

[0035] Community cloud: Cloud infrastructure is shared by several organizations and supports a specific community with common interests (e.g., mission, security requirements, policy, and compliance considerations). It can be managed by the organization or a third party and can exist on-premises or off-premises.

[0036] Public cloud: Cloud infrastructure is available to the public or to a large industry group and is owned by an organization that sells cloud services.

[0037] Hybrid cloud: A cloud infrastructure consisting of two or more clouds (private, community, or public) that remain distinct entities but are bound together by standardized or proprietary technologies that enable data and application portability (e.g., cloud burst sharing for load balancing between clouds).

[0038] The cloud computing environment is service-oriented and focuses on statelessness, low coupling, modularity, and semantic interoperability. The core of cloud computing is the infrastructure that consists of a network of interconnected nodes.

[0039] The cloud computing environment includes one or more cloud computing nodes, local computing devices used by cloud consumers, such as, for example, personal digital assistants (PDAs) or cell phones, desktop computers, laptop computers, etc. The nodes can communicate with each other. They can be physically or virtually grouped in one or more networks such as private clouds, community clouds, public clouds, or hybrid clouds, or a combination thereof (not shown). This allows the cloud computing environment to provide infrastructure, platforms, and / or software as a service, and for such services, cloud consumers do not need to maintain resources on local computing devices. It will be understood that the types of computing devices are for illustrative purposes only, and that computing nodes and cloud computing environments can communicate with any type of computerized device through any type of network and / or network addressable connection (e.g., using a web browser).

[0040] The cloud computing environment includes a set of functional abstraction layers. The following layers and corresponding functions are provided:

[0041] The hardware and software layers include hardware and software components. Examples of hardware components include: mainframes; servers based on RISC (Reduced Instruction Set Computer) architecture; servers; blade servers; storage devices; and network and networking components. In some embodiments, software components include network application server software and database software.

[0042] The virtualization layer provides an abstraction layer from which the following examples of virtual entities can be provided: virtual servers; virtual storage; virtual networks, including virtual private networks; virtual applications and operating systems; and virtual clients.

[0043] In one example, the management layer may provide the following functionality. Resource provisioning provides dynamic procurement of computing resources and other resources used to perform tasks within a cloud computing environment. Metering and pricing provides cost tracking when using resources within a cloud computing environment, as well as billing or invoicing for consumption of these resources. In one example, these resources may include application software licenses. Security provides authentication for cloud consumers and tasks, as well as protection of data and other resources. A user portal provides access to the cloud computing environment for consumers and system administrators. Service level management provides cloud computing resource allocation and management to meet required service levels. Service level agreement (SLA) planning and fulfillment provides pre-arrangement and procurement of cloud computing resources in anticipation of future demand based on the SLA.

[0044] The workload layer provides examples of functionality that can take advantage of a cloud computing environment. Examples of workloads and functionality that can be provided by this layer include: mapping and navigation; software development and lifecycle management; virtual classroom education delivery; data analytics processing; transaction processing; and digital report creation and failure event prediction and corrective action.

[0045] It should be understood that although the present disclosure includes detailed descriptions about cloud computing, the implementation of the teachings described herein is not limited to cloud computing environments. Rather, embodiments of the present invention can be implemented in conjunction with any other type of computing environment now known or later developed.

[0046] Systems and processes for converting controller data into digital reports

[0047] Figure 2 According to an embodiment of the present invention Figure 1 200 includes a controller data receiving module 202, a data conversion module 204, a digital report storage module 206, an extraction module 208, a GUI enhancement module 210, an AI training module 212, a simulation module 214, and a control module 216. The number of modules may vary, and some modules may be combined with other modules or separated into two or more modules. The functions of the modules included in the code 200 will be presented below. Figure 3 This is discussed in detail in the discussion of .

[0048] Figure 3 is a flow chart of a process for converting controller data into digital reports according to an embodiment of the present invention, wherein the operations of the flow chart are Figure 2 One or more modules in the . Figure 3 The processing starts from the starting node 300.

[0049] In step 302, the controller data receiving module 202 (see Figure 2) receives one or more data packets from a device controller via a network. In an embodiment, one or more data packets are received at fixed intervals (e.g., the device controller uploads a new data packet every hour). The device controller can be a controller or an onboard computer of the device, or one or more processors of a computing system associated with the device. The device can be an industrial or safety-critical device, such as a medical gas compressor system used in a hospital. The device can be a technical system that consists of one or more technical components, parts, systems, subsystems, etc. Examples of technical systems are compressor systems, multi-stage compressors, vacuum pumps, medical gas delivery systems, or any mechanical-based system that requires regular monitoring and / or maintenance. The one or more data packets include data associated with technical parameters of the device in a given state of the device, which data are from one or more sensors, alarm circuit systems, etc. of the device. The technical parameters include, but are not limited to, system pressure, ambient temperature, compressor exhaust temperature, pipeline pressure, speed of a variable speed drive, dew point, carbon monoxide level, carbon dioxide level, operating mode of a compressor component, operating hours of a component, average daily operating hours of a compressor, unit status, alarm status, voltage output, and voltage input. The technical parameters included in the digital report may vary by device, manufacturer, state and local laws, and hospital.

[0050] In step 304, the data conversion module 204 converts the data contained in the data packets into a digital report of a given condition of the compressor in a human readable format. The data contained in the one or more data packets are key-value pairs indicating a key associated with a technical parameter and a value associated with the technical parameter. Natural language processing tools and key-value matching can be used to convert the data into text for the digital report in a manner that semantically matches the organization's custom template so as to present the information as continuous natural language that can be parsed by conventional natural language processing algorithms and machine reading and comprehension engines.

[0051] In step 306, the digital report storage module 206 stores the digital report in one or more databases accessible through an online portal associated with the computer system. Organization members can log in from a local machine over the network at any time to view, retrieve, or otherwise access the digital report. In addition, members logged into the online platform can query the database(s) in natural language, and the report storage module 206 can retrieve the relevant digital report.

[0052] Systems and processes for predicting failure events using trained AI models

[0053] Figure 4 is a flowchart of a process for predicting a fault event and taking corrective action according to an embodiment of the present invention, wherein the operations of the flowchart are Figure 2 One or more modules in the . Figure 4The processing of starts at the start node 400 and ends at the end node 410.

[0054] In step 402 , the extraction module 208 uses the trained artificial intelligence model to extract predicted fault events and corrective actions of the compressor from the one or more data packets sent by the controller of the input compressor.

[0055] In step 404 , in response to the trained artificial intelligence model outputting a predicted fault event based on one or more data sets fed into the trained artificial intelligence model by one or more processors, the GUI enhancement module 210 enhances the graphical user interface of the controller to display a new icon indicating a potential fault event.

[0056] In step 406, the GUI enhancement module 210 blocks transmission of new data packets for the new state of the compressor from the controller of the compressor unless confirmation is received from the controller of the compressor that a physical input associated with the new icon of the graphical user interface has been received, and sends a notification to the online portal that transmission of new data packets is blocked.

[0057] Alternatively, the GUI enhancement module 210 accepts transmission of a new data packet for a new status of the compressor from the compressor controller to prepare a new digital report, wherein, upon said acceptance, one or more processors communicate with the controller to display on the graphical user interface that the transmission is blocked unless confirmation is received from the controller of the compressor that a physical input associated with the new icon of the graphical user interface was received by the controller.

[0058] In step 408, the AI ​​training module 214 feeds the output of the trained AI model into the training of the AI ​​model to continuously train the AI ​​model.

[0059] Systems and processes that simulate changes to technical systems using the output of trained AI models

[0060] The output of the trained AI model can be fed to a digital computer model of the compressor to simulate parameter adjustments of the compressor / medical air system before implementing adjustments to the active compressor located in the facility. In most cases, it is not ideal to remotely implement changes to the active compressor system because the active compressor system is being used by patients. Alternatively, a digital computer model of the compressor system is created to simulate changes and / or adjustments to the compressor system provided by the trained AI model. For example, the trained AI model can recommend changes to the active compressor system based on the analysis of the digital report by the trained AI model to improve the operation / function of the active compressor system. Before implementing changes to the system of the active compressor, the recommended changes to one or more operating parameters are simulated in the digital model of the compressor system. The simulation determines the impact of the recommended changes and can also determine the optimal parameters of the active compressor system. In an exemplary embodiment, the simulation can determine the most efficient way to operate the compressor / vacuum pump of the compressor system to minimize the operating time, thereby saving energy and wear of the compressor / pump. The simulation results can be accessed via an online portal for evaluation and implementation of the active compressor system.

[0061] Figure 5 is a flowchart of a process for simulating changes to a compressor system based on the output of a trained AI model according to an embodiment of the present invention, wherein the operations of the flowchart are Figure 2 One or more modules in the . Figure 5 The processing of starts at the start node 500 and ends at the end node 510.

[0062] In step 502, the extraction module 208 uses the trained artificial intelligence model to extract the operating parameters of the active compressor system from the one or more data packets sent by the compressor controller by inputting the data in the one or more data packets. The extraction module 208 feeds the digital report as input to the trained AI model, which extracts the operating / technical parameters of the active compressor system and evaluates the efficiency of the active compressor by comparing the one or more operating / technical parameters with one or more other operating / technical parameters of the active compressor. The output of the trained AI model is one or more recommended changes to the one or more operating / technical parameters of the active compressor system.

[0063] As an example, the trained AI model compares the amount of time each compressor is running based on the system's operating pressure to determine flow demand and determines that waiting for the operating pressure to decrease before turning on additional compressors will maintain the desired operating pressure while running fewer units. The trained AI model outputs a recommendation to lower the pressure band setting while maintaining the same system output pressure to reduce power consumption and excessive wear on the compressors. The trained AI model can analyze many different scenarios.

[0064] In step 504 , the simulation module 214 simulates the suggested change(s) suggested by the trained AI model in step 502 in the digital model of the technical system 145 .

[0065] In step 506, the simulation module 214 evaluates the impact of the change in the digital model 145. Continuing with the above example, the simulation module 214 reduces the pressure band setting while maintaining the same system output pressure and measures the impact on other operating / technical parameters of the digital model of the active compressor system. If the values ​​of the other operating / technical parameters are not affected by the change, then the simulation module 214 recommends the suggested change via the online portal. Conversely, if the values ​​of the other operating / technical parameters are adversely affected by the change, then the simulation module 214 does not recommend the suggested change and may provide a record in the online portal.

[0066] In step 508 , the AI ​​training module 214 feeds the results of the suggestions and / or simulations back into the trained AI model to continuously train the AI ​​model.

[0067] Systems and processes for controlling technology systems using the output of trained AI models

[0068] The output of the trained AI model is used to control active compressors located at the facility by implanting the recommended changes. The trained AI model can recommend changes to the active compressor system based on its analysis of the digital report to improve the operation / function of the active compressor system. One or more recommended changes to one or more operating parameters are implemented by sending a control signal to an equipment controller, which controls one or more physical components of the compressor system. The control signal is simulated in a digital model of the compressor system. A record of the changes can be accessed via an online portal.

[0069] Figure 6 is a flowchart of a process of controlling a compressor system based on the output of a trained AI model according to an embodiment of the present invention, wherein the operations of the flowchart are Figure 2 One or more modules in the . Figure 6 The processing of begins at the start node 600 and ends at the end node 612 .

[0070] In step 602, the extraction module 208 uses the trained artificial intelligence model to extract the operating parameters of the active compressor system from one or more data packets sent by the input compressor controller. The extraction module 208 feeds the digital report as input into the trained AI model, which extracts the operating / technical parameters of the active compressor system and evaluates the efficiency of the active compressor by comparing the one or more operating / technical parameters with one or more other operating / technical parameters of the active compressor. The output of the trained AI model is one or more recommended changes to the one or more operating / technical parameters of the active compressor system. As another example, the trained AI model compares the time between switching the drying tower and determines that the system does not need to purge as much air in order to still maintain the ability of the desiccant to absorb the required moisture. The trained AI model output recommends reducing the purge time of the drying tower. This causes the system to lose less compressed air, meaning that the compressors do not have to run as much as they do now, thereby saving energy consumed by the system and avoiding excessive wear on the compressor. The trained AI model can analyze many different scenarios.

[0071] In step 604 , the simulation module 214 simulates the suggested changes suggested by the trained AI model in step 502 in the digital model of the technical system 145 .

[0072] In step 606, the simulation module 214 evaluates the impact of the change to the digital model 145. Continuing with the above example, the simulation module 214 shortens the purge time of the drying tower and measures the impact on other operating / technical parameters of the digital model of the active compressor system. If the values ​​of the other operating / technical parameters are not affected by the change, then the simulation module 214 recommends the suggested change. Conversely, if the values ​​of the other operating / technical parameters are adversely affected by the change, then the simulation module 214 does not recommend the suggested change and may provide a record in the online portal.

[0073] In step 608, the control module 216 receives the recommended changes from the simulation module 214 and generates a control signal and sends the control signal to the equipment controller 180 via the network 140. The control signal is one or more electronic signals that instruct the equipment controller to make changes to the active compressor system. The equipment controller 180 sends control / command signals to one or more physical components (e.g., valves, actuators, drive units, onboard computers, etc.) of the active compressor system to change the technical parameters of the active component system.

[0074] In one example, to dry air for the active compressor system, it passes the air through alternating desiccant chambers. To dry the inactive chambers, some of the dry air passes through the desiccant and is purged to the equipment chamber to regenerate the desiccant. The trained AI model detects from the digital reports that the system appears to have little dryer activity and determines that additional actuators can be activated to prevent purging as much air as normal. Saving some of this air means that the compressors must run less frequently because they are not wasting some of the compressed air into the equipment chamber. Therefore, the control module 216, in response to the output of the trained AI model, sends a control signal to the equipment controller 180 to actuate the actuator responsible for purging the air. In addition, the trained AI model can learn that the timing of the purge cycle on the dryer can also be reduced, thereby reducing the purge time. The control module 216 can prepare a software update to push to the equipment controller 180 of the active compressor system, which will affect the timing variables of the dryer.

[0075] In another example, when the compressor is running, a valve is actuated at the output of the exhaust radiator to ensure that some of the moisture generated by the cooling of the air is periodically blown into the drain. This prevents excess moisture from entering the pressure vessel and potentially becoming humidity that needs to be removed by the dryer. The trained AI model determines from the digital reports that the system is not generating nearly as much heat because it is not running as frequently. Therefore, the trained AI model suggests that the timing and frequency of the actuator valve timing can be reduced so that less pressurized air is released from the equipment, which means that the compressor will run less time. The control module 216 sends a control signal to the equipment controller 180 to actuate the valve, thereby controlling the compressor system.

[0076] In another example, the starting and stopping of the compressor and vacuum pump (unit) is controlled by a pressure band, where the unit is turned on additionally when the pressure drops, and the unit is turned off when the pressure rises. In response to being fed the digital report, the trained AI model learns that if the pressure band is adjusted slightly to optimize the number of units that actually need to be started if the regular load can be handled by one less unit, then the customer's regulator flow will cause the system to turn units on and off more times than they would otherwise have to. The trained AI model can model different pressure bands based on the regular flow of the system to determine the optimal pressure band setting for the system. The control module 216 can implement the optimal pressure band setting for the system through the network 140 to automatically control the variables in the system. The new pressure band applied to the system can be analyzed to see if the change improved the system because the digital report will indicate that the average daily operating time of the compressor will be reduced.

[0077] In some embodiments, steps 604 and 606 (i.e., simulation) are optional, such that the recommended changes output by the trained AI model are not first simulated before being implemented into the active compressor system. If the confidence that the changes will not pose a threat to the operation of the active compressor is above a predetermined threshold, then the recommended changes can be implemented.

[0078] In step 610 , the AI ​​training module 214 feeds the applied changes back into the trained AI model to continuously train the AI ​​model.

[0079] Training AI models to predict component failures

[0080] The training module 212 of the code 200 uses supervised machine learning or unsupervised machine learning to train the first AI model. In an exemplary embodiment, the algorithm of the training model is to predict fault events and take corrective measures.

[0081] An input sample (e.g., a digital report) is fed into one or more neural networks to obtain a first vector and a second vector. An example of the first input sample is the volume of air purged from the system, and the second input sample is the average system pressure. Another example of the first input sample is an alarm condition, and the second input sample is the temperature of the system. Another example of the first input sample is the moisture content associated with the discharge radiator, and the second input sample is the pressure associated with the discharge radiator. A number between 0 and 1 is calculated using a cosine similarity function of the first vector and the second vector. Numbers closer to 1 indicate a match between sample types, and numbers closer to 0 indicate that the sample types do not match. A knowledge graph using the relative relationship of at least two inputs is constructed and translated into readable text to understand the impact on the life of the compressor system components. The inputs can be represented on the knowledge graph together with their relative relationships to each other. The knowledge graph is translated into a machine-readable natural language using a natural language generation method based on knowledge graph data in a graphical form to organize the semantic information of each node in the knowledge graph into a continuous natural language temporary text, and then translated into a machine-readable natural language using a natural language style conversion method. In this way, the second AI model is trained to learn which components are likely to fail, if any, within a predefined time window.

[0082] Training AI models to simulate and / or control compressor systems

[0083] The training module 212 of the code 200 trains the first AI model using supervised machine learning or unsupervised machine learning. In an exemplary embodiment, the algorithm of the training model determines how the compressor system can be controlled to improve efficiency and / or avoid failure.

[0084] An input sample (e.g., a digital report) is fed into one or more neural networks to obtain a first vector and a second vector. An example of the first input sample is the average daily operating time of a vacuum pump, and the second input sample is the average system pressure. Another example of the first input sample is the speed of a variable speed drive, and the second input sample is the average daily operating time of a variable speed drive. Another example of the first input sample is the moisture content associated with a discharge radiator, and the second input sample is the temperature associated with the discharge radiator. A number between 0 and 1 is calculated using a cosine similarity function of the first vector and the second vector. Numbers closer to 1 indicate a match between sample types, and numbers closer to 0 indicate that the sample types do not match. A knowledge graph using the relative relationship of at least two inputs is constructed and translated into readable text to understand the impact on the life of compressor system components. The inputs can be represented on the knowledge graph together with their relative relationships to each other. The knowledge graph is translated into a machine-readable natural language using a natural language generation method based on knowledge graph data in graph form to organize the semantic information of each node in the knowledge graph into a continuous natural language temporary text, and then translated into a machine-readable natural language using a natural language style conversion method. In this way, the second AI model is trained to learn what corrective actions, if any, can be taken to improve the efficiency of the compressor system.

[0085] The description of various embodiments of the present invention is for illustrative purposes only and is not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements over technologies found on the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A computer-implemented method comprising: receiving, by one or more processors of the computer system, one or more data packets from a controller of the compressor over a network, the one or more data packets including data associated with a technical parameter of the compressor at a given state of the compressor; converting, by the one or more processors, the data contained in the data packet into a digital report of a given condition of the compressor in a human-readable format; as well as The digital reports are stored by the one or more processors in one or more databases accessible through an online portal associated with the computer system.

2. The computer-implemented method of claim 1 , further comprising: The one or more processors are configured to input data in the one or more data packets sent by the controller of the compressor and extract predicted fault events and corrective measures for the compressor from the one or more data packets using the trained artificial intelligence model.

3. The computer-implemented method of claim 1 , further comprising: training, by the one or more processors, a trained artificial intelligence model to output predicted fault events of the compressor and corrective measures to avoid the predicted fault events by feeding an input data set of actual fault events of the compressor and known solutions to avoid or correct the actual fault events into the artificial intelligence model; constructing, by the one or more processors, a knowledge graph using a relative relationship of at least two nodes of an input data set of actual fault events; as well as The knowledge graph is translated into a machine-readable natural language by the one or more processors.

4. The computer-implemented method of claim 1, wherein the controller includes a graphical user interface that allows transmission of the one or more data packets to be received by the computer system over the network only in response to physical input.

5. The computer-implemented method of claim 4, further comprising: The one or more processors, in response to the trained artificial intelligence model outputting a predicted fault event based on the one or more data sets fed into the trained artificial intelligence model by the one or more processors, enhance the controller's graphical user interface to display a new icon indicative of a potential fault event.

6. The computer-implemented method of claim 5, further comprising: preventing, by the one or more processors, transmission of a new data packet from a controller of the compressor for a new state of the compressor unless confirmation is received from the controller of the compressor that a physical input associated with a new icon of the graphical user interface was received; as well as A notification is sent, by the one or more processors, to the online portal that transmission of the new data packet is blocked.

7. The computer-implemented method of claim 5, further comprising: and accepting, by the one or more processors, transmission of a new data packet for a new state of the compressor from a controller of the compressor to prepare a new digital report, wherein, upon the acceptance, the one or more processors communicate with the controller to display on the graphical user interface that the transmission is blocked unless confirmation is received from the controller of the compressor that a physical input associated with a new icon of the graphical user interface was received by the controller.

8. The computer-implemented method of claim 1 , wherein converting the data in the one or more data packets into a digital report comprises: Format from one data file type to another.

9. The computer-implemented method of claim 1, wherein the compressor technical parameters include system pressure, ambient temperature, compressor discharge temperature, line pressure, variable speed drive speed, dew point, carbon monoxide level, carbon dioxide level, operating mode of compressor components, operating hours of components, average daily operating hours of the compressor, unit status, alarm status, voltage output, and voltage input.

10. The computer-implemented method of claim 1, wherein the data contained in the one or more data packets are key-value pairs indicating a key associated with a technical parameter and a value associated with the technical parameter.

11. The computer implemented method of claim 1, wherein the one or more data packets are received at predetermined intervals and represent a new state of the compressor.

12. The computer-implemented method of claim 1 , further comprising: The text of the digital report is reconstructed by the one or more processors to present the information as continuous natural language capable of being parsed by conventional natural language processing algorithms and machine reading and comprehension engines.

13. A computer program product comprising a computer-readable hardware storage medium embodied with program instructions, the program instructions being readable by one or more processors of a computer system to cause the one or more processors to: receiving one or more data packets from a controller of the compressor via a network, the one or more data packets including data associated with a technical parameter of the compressor at a given state of the compressor; converting the data contained in the data packet into a digital report of a given condition of the compressor in a human-readable format; and The digital reports are stored in one or more databases accessible through an online portal associated with the computer system.

14. A computer system comprising: one or more computer processors; one or more computer-readable storage media; as well as Computer readable code collectively stored in the one or more computer readable storage media, the computer readable code including data and instructions to cause the one or more computer processors to perform at least the following operations: receiving, by one or more processors of the computer system, one or more data packets from a controller of the compressor over a network, the one or more data packets including data associated with a technical parameter of the compressor at a given state of the compressor; converting, by the one or more processors, the data contained in the data packet into a digital report of a given condition of the compressor in a human-readable format; as well as The digital reports are stored by the one or more processors in one or more databases accessible through an online portal associated with the computer system.

15. The computer system of claim 14, further comprising: The one or more processors input data in the one or more data packets sent by the controller of the compressor and use the trained artificial intelligence model to extract predicted fault events of the compressor from the one or more data packets.

16. The computer system of claim 14, further comprising: training, by the one or more processors, a trained artificial intelligence model to output predicted fault events of the compressor and corrective measures to avoid the predicted fault events by feeding an input data set of actual fault events of the compressor and known solutions to avoid or correct the actual fault events into the artificial intelligence model; constructing, by the one or more processors, a knowledge graph using a relative relationship of at least two nodes of an input data set of actual fault events; as well as The knowledge graph is translated into a machine-readable natural language by the one or more processors.

17. The computer system of claim 14, wherein the controller includes a graphical user interface that allows transmission of the one or more data packets to be received by the computer system over the network only in response to physical input.

18. The computer system of claim 17, further comprising: The one or more processors, in response to the trained artificial intelligence model outputting a predicted fault event based on the one or more data sets fed into the trained artificial intelligence model by the one or more processors, enhance the controller's graphical user interface to display a new icon indicative of a potential fault event.

19. The computer system of claim 18, further comprising: preventing, by the one or more processors, transmission of a new data packet from a controller of the compressor for a new state of the compressor unless confirmation is received from the controller of the compressor that a physical input associated with a new icon of the graphical user interface was received; as well as A notification is sent, by the one or more processors, to the online portal that transmission of the new data packet is blocked.

20. The computer system of claim 19, further comprising: and accepting, by the one or more processors, transmission of a new data packet for a new state of the compressor from a controller of the compressor to prepare a new digital report, wherein, upon the acceptance, the one or more processors communicate with the controller to display on the graphical user interface that the transmission is blocked unless confirmation is received from the controller of the compressor that a physical input associated with a new icon of the graphical user interface was received by the controller.

Images