Techniques for performance diagnostics of analytical instruments

Abstract

Techniques and apparatuses for diagnostic processes of analytical instruments are described. In one embodiment, for example, an apparatus can include at least one memory and logic coupled to the at least one memory. The logic can be configured to receive diagnostic information associated with at least one analytical instrument and process the diagnostic information using a computational model to generate at least one diagnostic model for at least one diagnostic device. Other embodiments are described.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 62 / 859,565, filed June 10, 2019, entitled “TECHNIQUES FOR ANALYTICALINSTRUMENT PERFORMANCE DIAGNOSTICS,” which is hereby incorporated by reference. Technical Field

[0003]

[0014] Embodiments herein relate generally to managing analytical instruments, and more particularly to processes for monitoring analytical instrument performance and diagnosing abnormal operating conditions of analytical instruments. Background Art

[0004] The performance of analytical instruments is continuously monitored to ensure data quality. Analytical instruments such as mass spectrometry (MS) and / or liquid chromatography-mass spectrometry (LC-MS) systems are capable of providing detailed characterization of complex sample sets. However, the ability to perform analyses to obtain precise, detailed analytical data makes MS and LC-MS systems susceptible to operational instabilities, including hardware and / or software problems that are difficult to detect and diagnose. Such problems may be caused by existing or impending component failure, human error, incorrect system configuration, etc. In conventional systems, determining the root cause often involves an extensive and time-consuming trial-and-error process performed by operators who may have experience running methods on the analytical instrument, but may not have sufficient knowledge of the specific system components to efficiently determine the diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 An embodiment of a first operating environment according to an embodiment is shown.

[0006] Figure 2 An embodiment of a second operating environment is shown.

[0007] Figure 3 An embodiment of a third operating environment is shown.

[0008] Figure 4 A configuration structure for diagnosis is shown.

[0009] Figures 5 to 16 Shown are diagrams and associated data capture files for various diagnostic scenarios according to some embodiments.

[0010] Figures 17 to 39 An illustrative diagnostic scenario for a chromatography system is shown, according to some embodiments.

[0011] Figure 40An embodiment of a computing architecture is shown. DETAILED DESCRIPTION

[0012] Various embodiments may generally relate to systems, methods, and / or devices for monitoring the performance of an analytical instrument. In some embodiments, a diagnostic service process may be used to determine a standard operating condition of an analytical instrument and / or its components. In various embodiments, the standard operating condition may include analytical instrument operations associated with standard reference data (e.g., data corresponding to normal, expected, and / or similar operations of the analytical instrument). In an exemplary embodiment, a diagnostic service process may be operable to determine a non-standard operating condition of an analytical instrument and / or its components. In various embodiments, the non-standard operating condition may include analytical instrument operations associated with non-standard reference data (e.g., data corresponding to abnormal conditions, unexpected conditions, component failure (or impending failure) conditions, instrument error conditions, and / or similar operations of the analytical instrument). In some embodiments, the diagnostic service process may be operable to determine the cause of the non-standard operating condition.

[0013] In various embodiments, data generated during the operation of an analytical instrument and / or its components can be accessed, for example, via data channels. Some embodiments can implement automated computer software with machine learning using neural networks and big data (e.g., large data sets that can be computationally analyzed to reveal patterns, trends, and correlations between data channels). The diagnostic service process can be operated to perform a training process using the data. Non-limiting examples of the training process can include developing a pass / fail input / output function to determine the behavior of a given instrument relative to a pool of recorded repeated positive / negative tests occurring during normal operation of the instrument / system (e.g., initialization, flow, or run injection of a liquid chromatography-mass spectrometry (LC-MS) system). The diagnostic service process can be operated to perform a testing process, such as detecting and / or diagnosing instrument issues (e.g., non-standard operating conditions) with an analytical instrument and / or its components. In various embodiments, the testing process can include applying the pass / fail input / output function to predetermined tests (e.g., tests determined during the training process) and providing the collected results and data to a training data repository.

[0014] In various embodiments, analytical instrument and / or its components can generate operational data in various possible scenarios, including but not limited to when stationary, rapid diagnostic test, or when in normal operation (e.g., performing analytical methods). In an exemplary embodiment, operational data can be or may include time-based digital binary analog data from internal devices. In some embodiments, channel data can be subscribed and read via logical components, processors, controllers, circuits, etc. residing on a computing device that is operably coupled to analytical instrument and / or its components. In some embodiments, the system may include one or more analytical instruments, its components and / or related equipment (e.g., sampling equipment, reagent supply equipment, etc.). Each system may include one or more logical components (or controllers) and / or be associated therewith.

[0015] In some embodiments, the logic component may receive configuration information for analytical instruments and / or their components. In various embodiments, the configuration information may include, but is not limited to, operating configurations, modes, ranges, parameters, thresholds, and the like. In some embodiments, the logic component may receive configuration data items and execute a training process or pattern using the configuration data, channel data, and associated diagnostics stored in a data repository and input into one or more computational models (e.g., neural networks) to create models for each instrument, system, device, component, etc. and / or its operating mode, configuration, etc. In some embodiments, the logic component may receive configuration data items and execute a testing process or pattern using the models for each instrument, system, device, component, etc. and / or its operating mode, configuration, etc., using the configuration data and real-time channel data input into one or more computational models (e.g., neural networks) to determine one or more diagnoses. In various embodiments, the test channel data and / or diagnostic information may be added to the training data repository. In some embodiments, the diagnostics for the instruments, systems, devices, components, etc. and / or their operating mode, configuration, etc. may be provided to operators for review, analysis, annotation (or "flagging"), etc.

[0016] In some embodiments, the diagnostic service process may include a machine learning framework of the form y = f(x), where y is the output, f is the prediction function, and x is the input. For example, during training, given a training set of labeled instances {(x1, y1), …, (xN, yN)}, the prediction function f can be estimated to minimize the prediction error on the training set. For another example, during testing, f can be applied to a new test instance x, and the output prediction value y = f(x) can be determined.

[0017] In various embodiments, a computational model can be generated for an analytical system, which may include an analytical instrument, analytical instrument components, and / or associated equipment. The computational model may include a standard model and / or model data associated with standard operation (e.g., normal operation, expected operation, appropriate operation, operation within a range or threshold, etc.), wherein the system and / or its components (e.g., analytical instrument components and / or associated equipment) operate properly, such as within acceptable limits. The computational model may also include a non-standard model and / or non-standard model data associated with non-standard operation, wherein the system and / or its components (e.g., analytical instrument components and / or associated equipment) operate incorrectly, such as outside acceptable limits. In this way, system failures and maintenance can be predicted through predictive modeling of the system and / or its components. Thus, diagnostic service processes according to some embodiments can determine when a system and / or system component has failed and / or predict when a system and / or system component will fail, and address such issues earlier and with greater accuracy than conventional systems that rely on trial-and-error techniques implemented by an operator. Consequently, systems operating according to some embodiments can achieve higher levels of operational efficiency compared to conventional systems.

[0018] Some embodiments may operate the diagnostic process at various levels. For example, the levels may include at the unit level, which may involve low-level discrete electromechanical sensing, such as motors, solenoids, columns, etc. In another example, the levels may include at the component level, which may involve subassembly components, such as pumps, heater coolers, sample handling devices, etc. In another example, the levels may include at the qualification level, which may include at the functional domain level, such as product initialization, instrument qualification, instrument testing, etc. In other examples, the levels may include at the comprehensive level, which may involve at the application level representing the system, which may include any post-processing data (e.g., monitoring instrument data (raw / processed) to correlate patterns and trends with specific hardware or method issues).

[0019] In this description, numerous specific details, such as components and system configurations, may be set forth in order to provide a more thorough understanding of the described embodiments. However, those skilled in the art will appreciate that the described embodiments may be practiced without such specific details. Additionally, some well-known structures, elements, and other features have not been shown in detail to avoid unnecessarily obscuring the described embodiments.

[0020] In the following description, references to "one embodiment," "an embodiment," "an exemplary embodiment," "various embodiments," etc., indicate that an embodiment of the described technology may include a particular feature, structure, or characteristic, but more than one embodiment may include the particular feature, structure, or characteristic and not every embodiment must include the particular feature, structure, or characteristic. Furthermore, some embodiments may have some, all, or none of the features described for other embodiments.

[0021] As used in this specification and the claims, unless otherwise indicated, the use of ordinal adjectives "first," "second," "third," etc. to describe elements merely refers to a specific instance of the element or different instances of similar elements, and is not meant to imply that the elements so described must be in a specific order in time, space, ranking, or in any other manner.

[0022] Figure 1 An example of an operating environment 100 is shown that may represent some embodiments. Figure 1 As shown, the operating environment 100 may include a system 105 for managing information associated with devices 115a to 115n. In some embodiments, the devices 115a to 115n may be or may include analytical instruments, components thereof (e.g., pumps, valves, instrument modules, etc.), equipment associated therewith (e.g., sample management systems, reagent systems, etc.), etc. In some embodiments, the analytical instruments may be or may include chromatography systems, liquid chromatography (LC) systems, gas chromatography (GC) systems, mass analyzer systems, mass detector systems, mass spectrometer (MS) systems, ion mobility spectrometer (IMS) systems, high performance liquid chromatography (HPLC) systems, ultra high performance liquid chromatography (UPLC) systems, and the like. ® ) systems, ultra high performance liquid chromatography (UHPLC) systems, ultraviolet (UV) detectors, visible light detectors, solid phase extraction systems, sample preparation systems, sample introduction systems, pump systems, capillary electrophoresis instruments, combinations thereof, components thereof, variations thereof, and the like. Although LC, MS, and LC-MS are used in the examples in this detailed description, the embodiments are not limited thereto, as other analytical instruments capable of operating according to some embodiments are contemplated herein.

[0023] In various embodiments, devices 115a to 115n may be or include various components, modules, and the like. Non-limiting examples of components may include detectors, solvent managers, sample managers, pumps, valves, electrodes, quadrupole elements, columns, ion source devices, sensors (e.g., pressure sensors, temperature sensors, flow sensors, etc.), data system hardware, data system software, and the like. In various embodiments, device components may include any component that can be individually monitored (e.g., associated with operational information). In some embodiments, devices 115a to 115n may be or include instrument systems. Generally speaking, an instrument system may include a collection of instrument modules that operate together as a cohesive unit. For example, an instrument system may include an analytical device (e.g., an LC or MS device) capable of providing analytical data (measurements) and supporting components that facilitate the generation of analytical data by the analytical device (e.g., injectors, detectors, pumps, control systems, etc. for the LC or MS system). The embodiments are not limited in this context.

[0024] In some embodiments, devices 115a to 115n are operable to perform analysis and generate analytical information 144. In various embodiments, analytical information 144 may include information, data, files, charts, graphs, images, textual information, and the like generated by the analytical instrument as a result of performing an analytical method. For example, in an LC-MS system, devices 115a to 115n may separate a sample according to a specific method and perform mass analysis on the separated sample to generate analytical information 144. This analytical information may include raw, original, or other unprocessed data, chromatograms, spectra, peak lists, mass values, retention time values, concentration values, compound identification information, and the like. In various embodiments, analytical information 144 may include information generated by quality control processes (such as system suitability testing, quality control checks, and the like). In some embodiments, system 105 may include multiple devices 115a to 115n. In various embodiments, system 105 may include a single device 115a to 115n.

[0025] In various embodiments, system 105 may include a computing device 110 communicatively coupled to devices 115a-115n or otherwise configured to receive and store diagnostic information associated with devices 115a-115n (e.g., in a data repository 154a-154n or other remote data storage structure, including a cloud storage environment). Generally speaking, diagnostic information may include information associated with devices 115a-115n that can be used to generate a diagnosis associated with devices 115a-115n. In some embodiments, diagnostic information may include analytical information 144 and / or operational information 146. Analytical information 144 may include information generated by devices 115a-115n as a result of performing an analysis. For example, for an LC-MS device, analytical information 144 may include, but is not limited to, compound lists, peaks, peak lists, quality control (QC) results, spectra, graphs, charts, and the like. In various embodiments, operational information 146 may include information associated with the operation of devices 115a-115n. For example, operational information 146 may include the voltage of a pump used by the LC-MS device during analysis. Non-limiting examples of operational information 146 may include voltage, current, power, flow information (e.g., flow rate, blockage detection, etc.), component position (e.g., valve open / closed position), alarms, alarm codes, vibration detection, audio detection, video images, pressure, temperature, software status, hardware status, etc. In some embodiments, operational information 146 may be obtained via one or more data channels 117a to 117n associated with devices 115a to 115n, their components, their sensors (e.g., pressure transducers associated with pumps, position sensors associated with valves, video recorders associated with components, etc.). Embodiments are not limited in this context.

[0026] In various embodiments, the devices 115a-115n and / or the data channels 117a-117n are operable to provide the analysis information 144 and / or the operational information 146 directly to the computing device 110 and / or a location on the network 150 accessible to the computing device 110, such as a cloud computing environment, nodes 152a-152n, data repositories 154a-154n, etc. In various embodiments, the data channels 117a-117n may operate with various forms of data, including, but not limited to, numeric data, alphanumeric data, graphical data, textual data, visual data, audio data, image data, binary data, etc. The embodiments are not limited in this context.

[0027] In some embodiments, computing device 110 may be used to control, monitor, manage, or otherwise handle various operational functions of devices 115a through 115n. In some embodiments, computing device 110 may be used to provide analytical information 144 and / or operational information to locations on network 150 via a secure or authenticated connection. In some embodiments, computing device 110 may be or include a standalone computing device, such as a personal computer (PC), a server, a tablet computing device, a cloud computing device, a mobile computing device (e.g., a smartphone, a tablet computing device, etc.), a data appliance, etc. In various embodiments, computing device 110 may be or include a controller or control system integrated into devices 115a through 115n to control operational aspects thereof.

[0028] although Figure 1 Only one computing device 110 is depicted, but embodiments are not limited thereto. In various embodiments, the functions, operations, configurations, data storage functions, applications, logic, etc. described with respect to computing device 110 may be performed by and / or stored in one or more other computing devices. For illustrative purposes only, a single computing device 110 is depicted to simplify the drawing.

[0029] Computing device 110 may include processor circuit 120, memory unit 130, and transceiver 160. Processor circuit 120 may be communicatively coupled to memory unit 130 and / or transceiver 160.

[0030] The processor circuit 120 may include and / or have access to various logic for performing processes according to some embodiments. For example, the processor circuit 120 may include and / or have access to device model logic 124 and / or diagnostic logic 126. The processing circuit 120 and / or device model logic 124 and / or diagnostic logic 126 and / or portions thereof may be implemented in hardware, software, or a combination thereof. As used in this application, the terms "logic," "component," "layer," "system," "circuit," "decoder," "encoder," and / or "module" are intended to refer to a computer-related entity, which may be hardware, a combination of hardware and software, software, or software in execution, examples of which are provided by the exemplary computing architecture 4000. For example, logic, circuitry, controllers, processors, etc. may be and / or may include, but are not limited to, a process running on a processor, a processor, a hard drive, multiple storage drives (optical and / or magnetic storage media), an object, an executable program, a thread of execution, a program, a computer, a hardware circuit, an integrated circuit, an application specific integrated circuit (ASIC), a programmable logic device (PLD), a digital signal processor (DSP), a field programmable gate array (FPGA), a system on a chip (SoC), a memory cell, a logic gate, a register, a semiconductor device, a chip, a microchip, a chipset, a software component, a program, an application, firmware, a software module, computer code, any combination of the foregoing, etc.

[0031] although Figure 1 , diagnostic service logic 122 is depicted as being within processor circuitry 120, but embodiments are not limited thereto. Furthermore, while device model logic 124 and diagnostic logic 126 are depicted as logic within diagnostic service logic 122, embodiments are not limited thereto, as device model logic 124 and diagnostic logic 126 may be separate and / or independent logic. For example, diagnostic service logic 122 and / or any component thereof may be located within an accelerator, a processor core, an interface, a single processor die, implemented entirely as a software application (e.g., analysis service application 150), and so forth.

[0032] The memory unit 130 may include various types of computer-readable storage media and / or systems in the form of one or more higher-speed memory units, such as read-only memory (ROM), random-access memory (RAM), dynamic RAM (DRAM), double-data-rate DRAM (DDRAM), synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, polymer memory (such as ferroelectric polymer memory, bidirectional memory, phase change or ferroelectric memory), silicon-oxide-nitride-oxide-silicon (SONOS) memory, magnetic or optical cards, device arrays (such as redundant array of independent disks (RAID) drives), solid-state memory devices (e.g., USB memory, solid-state drives (SSDs), and any other type of storage medium suitable for storing information. In addition, the memory unit 130 may include various types of computer-readable storage media in the form of one or more lower-speed memory units, including internal (or external) hard disk drives (HDDs), magnetic floppy disk drives (FDDs), optical disk drives for reading from or writing to removable optical disks (e.g., CD-ROMs or DVDs), solid-state drives (SSDs), and the like.

[0033] The memory unit 130 can store an analysis service application 150 that can operate alone or in combination with the diagnostic service logic 122 to perform various diagnostic process functions according to some embodiments. In various embodiments, the analysis service application 150 can interact with the devices 115a to 115n and / or their components through various drivers (e.g., application programming interfaces (APIs)), software and / or hardware interfaces, etc.

[0034] In various embodiments, the diagnostic service logic 122 may be configured to provide and / or implement diagnostic services for the devices 115a-115n. In some embodiments, the diagnostic services may include processes for determining whether the devices 115a-115n and / or their components are operating normally or abnormally. In various embodiments, the diagnostic services may include processes for predictive maintenance, for example, determining whether the devices 115a-115n and / or their components are likely to operate abnormally in the future (e.g., to detect an impending fault condition). In an exemplary embodiment, the diagnostic services may include processes for diagnosing the source of the abnormal operation (e.g., a faulty pump, an inadequate seal, a degraded reagent, etc.). The embodiments are not limited in this context.

[0035] Device model logic 124 is operable to generate models associated with devices 115a to 115n and / or their components. In some embodiments, the models may be stored as diagnostic models 142. In various embodiments, the models may include various computational models, including but not limited to neural network models. Non-limiting examples of computational models may include machine learning models, artificial intelligence (AI) models, neural networks (NNs), artificial neural networks (ANNs), convolutional neural networks (CNNs), deep learning (DL) networks, deep neural networks (DNNs), recurrent neural networks (RNNs), combinations thereof, variations thereof, and the like. Embodiments are not limited in this context. In some embodiments, each diagnostic model 142 may be associated with at least one diagnostic device.

[0036] In some embodiments, the diagnostic model 142 may be or may include a neural network. For example, Figure 2 An example of an operating environment 200 is shown that can represent various embodiments of implementing a neural network 225. While neural networks may be used as examples herein, embodiments are not limited thereto, as any type of model capable of operating according to some embodiments is contemplated herein. Generally speaking, the neural network 225 may be formed from multiple layers, including an input layer 250, hidden layers 260a through 260n, and an output layer 270. Each layer of the neural network 225 may include one or more neurons 280a through 280n. Generally, neurons 280a through 280n may be computational units that take one or more inputs, apply a function (e.g., an activation function) to the inputs, and generate an output. Neurons 280a through 280n may be interconnected such that the output from neurons 280a through 280n in a first layer may be input to neurons 280a through 280n in a subsequent layer (and vice versa). For example, the output of neuron 280a in the input layer 250 may be provided as input to neurons 280d through 280f in the hidden layer 260a. Generally speaking, input layer 250 and output layer 270 are visible outside of neural network 225 (e.g., available as network input or accessible as network output, respectively), such as to software, an agent, etc. Hidden layers 260 a - 260 n are not accessible from outside neural network 225 and operate as hidden (or deep) processing structures for processing (e.g., training and / or analyzing) input data received from input layer 250 .

[0037] The results of the training or analysis can be provided to the output layer 280 and transmitted as output 230 to, for example, a data consumer. For example, in some embodiments, the diagnostic service process may include a neural network having the form y = f(x), where y is the output, f is the prediction function, and x is the input. For example, during the training process, given a training set of labeled instances {(x1, y1), …, (xN, yN)}, the neural network 225 can implement the prediction function f by minimizing the prediction error on the training set. For another example, during the testing process, the neural network can apply f to a new test instance x and output a predicted value y = f(x). The embodiments are not limited in this context.

[0038] In some embodiments, the device model logic 124 may generate and train the model based on training information 140. In various embodiments, the training information 140 may be generated by an operator (e.g., manual operational data), by operating the devices 115a through 115n and / or their components, or a combination thereof. For example, an operator may provide training information 140 generated during actual operation of a device component (such as a pump) to the computing device 110 as training information 140 (e.g., trusted operational data). In another example, the devices 115a through 115n and / or their components may be operated to generate the training information 140. In some embodiments, the operational information 140 may be generated for normal and / or abnormal operation of the devices 115a through 115n and / or their components. In one example, an operator may provide operational information 140 that includes channel data associated with the operation of a pump under normal conditions. In another example, an operator may provide operational information that includes channel data associated with abnormal operation of the pump (such as a low voltage condition). Depending on the operational condition, the operational information 140 may be designated as normal or abnormal.

[0039] Device model logic 124 can apply computational models, learning processes, and the like to training information 140 to train a diagnostic model. In various embodiments, a diagnostic model can be generated for one or more diagnostic devices. Generally speaking, a diagnostic device or diagnostic unit can include an entire device or system (e.g., an instrument cluster) or at least one monitored unit of devices 115a to 115n, channels 117a to 117n, their components, their functions, their characteristics, and the like. For example, a diagnostic device can include devices 115a to 115n, such as LC-MS devices. In another example, a diagnostic device can include components of devices 115a to 115n, such as columns, pumps, and valves. In another example, a diagnostic device can include the functions of devices 115a to 115n and / or their components, such as component operation (e.g., valve opening / closing, pump operation), and the like. In yet another example, a diagnostic device can include characteristics of devices 115a to 115n and / or their components, such as temperature, pressure, voltage, state (e.g., on / off, open / closed, etc.), and the like. In various embodiments, the diagnostic model logic 124 may continuously train the diagnostic model 142 with new information to improve the predictive capabilities of the diagnostic device and the diagnostic model 142 (eg, enabling a “big data” data repository that enables the diagnostic model 142 to become more accurate).

[0040] In some embodiments, the diagnostic logic 126 may, for example, test the diagnostic model 142 against actual data (e.g., actual channel data) to determine a diagnosis 148, which may include a prediction, an operational condition, and the like. For example, the diagnostic logic may implement y=f(x), where y is the output (e.g., the diagnosis 148), f is the prediction function (e.g., the diagnostic model 142), and x is the input (the operational information 146 in the form of actual channel data). For example, data channel 117a may include voltage operational information 146 for a pump of device 115a. The diagnostic logic 126 may receive the actual data as input via data channel 117a using the diagnostic model 142 associated with the pump and / or data channel 117a, and may generate a diagnosis regarding whether the pump is operating properly and / or make a diagnosis 148, including a prediction regarding the operational life of the pump (e.g., there is an X% probability of failure within a specific time period).

[0041] For example, for system 1 comprising devices 1-3, diagnosis 148 may include the following: system 1 is 100% normal, device 1 is 100%, device 2 is 100%, and device 3 is 100%. Diagnosis 148 for system 1 may indicate that system 1 is operating at 100% normal, and devices 1-3 are operating at 100%. For another example, diagnosis 148 for system 2 comprising devices 4-6 may include the following: system 2 is 66% normal, device 4 is 0%, device 5 is 100%, and device 6 is 100%. Diagnosis 148 for system 2 may indicate that system 2 is 66% normal, and device 4 is operating at 0%. Diagnosis 148 may further predict or indicate that device 4 may be the cause of system 2's 66% operating condition. In various embodiments, operational information 146 used to generate diagnosis 148 for system 1 and / or system 2 may be added to training information 140 to increase the amount of data available for generating diagnostic model 142 (e.g., achieving "big data" to increase model accuracy).

[0042] Machine learning in an analytical environment according to some embodiments can provide multiple technical advantages over conventional systems. One non-limiting technical advantage can include increasing the speed and accuracy of instrument / system diagnostics to many times that achievable by expert users. Another non-limiting technical advantage can include that, after training on an initial training data set, the diagnostic model 142 can be continuously and automatically updated at startup / setup intervals and improved based on additional capture (e.g., at customer sites) using "big data" that can be computationally analyzed to reveal patterns, trends, and correlations that may not be effectively and / or accurately determined through conventional operator methods.

[0043] Figure 3 An embodiment of the third operating environment is shown. Figure 3 As shown, operating environment 300 may include multiple instrument systems 315a to 315n communicatively coupled to system controller 320a via system diagnostic channels 317. In some embodiments, multiple diagnostic channels 317 may be used to inspect and diagnose internal devices. In various embodiments, there may be one diagnostic channel 317 for each monitored component, such as one diagnostic channel for each sensor. For example, for an LC-MS system, one or more diagnostic channels 317 may be generated via a test / injection, etc.

[0044] In some embodiments, the system controller 320 may include a processing configuration 322 and a processor 324. In various embodiments, there may be one system controller 320 for each instrument system 315a to 315n. In various other embodiments, the system controller 320 may interact with multiple instrument systems 315a to 315n. According to some embodiments, the processor 320 may implement various modeling processes (e.g., via machine learning processes). For example, the processor 320 may implement a training mode and / or a testing mode. In some embodiments, the training mode may include using configuration data, specific channel data, and related diagnostics stored in a data repository (e.g., a local data repository, a cloud data repository, etc.) as input into a neural network diagnostic model to create a diagnostic model for a diagnostic unit (e.g., an instrument, a system (e.g., a group of instruments and / or instrument components), a device, a component, a sensor, an element, etc.). In some embodiments, the training mode may include developing a pass / fail input / output function via an automated machine learning process to determine the behavior of a given instrument and / or its components relative to a pool of many recorded repeated positive / negative tests (e.g., discrete, simple tests) occurring during normal operation of the instrument / system (e.g., initialization, flow, injection, etc.).

[0045] In various embodiments, the test mode may include predicting a diagnosis of the diagnostic unit 326a to 326n using configuration data and actual (e.g., real-time or substantially real-time) channel data input into an existing neural network diagnostic model of the diagnostic unit. In an exemplary embodiment, the test mode may include diagnosing a problem with a given instrument and / or its components after applying a previously obtained pass / fail input / output function to a positive / negative test of the training mode, and adding the collected results and data to the training pool.

[0046] In some embodiments, diagnoses 326a through 326n may be associated with diagnostic units and diagnostic models. For example, diagnosis 326a may be associated with a first diagnostic device (e.g., the entire instrument system) and a specific neural network (neural network 1) generated for the first diagnostic device. In another example, diagnosis 326b may be associated with a second diagnostic device (e.g., a specific device in the instrument system, such as a pump) and a second neural network (neural network 2) associated with the second diagnostic device. For diagnosis 326a, processor 324 may receive channel data associated with the first diagnostic device and provide the channel data to the corresponding model (e.g., neural network 1) to generate diagnosis 326a.

[0047] In an exemplary embodiment, processed diagnoses for the instrument systems 315a-315n (e.g., the entire instrument system and each device in the instrument system) can be reported to the instrument system (e.g., in the following format: instrument system A 315a is 100% healthy, device 1 is 100%, device 2 is 100%, device 3 is 100%; and instrument system B is 66% healthy, device 4 is 0%, device 5 is 100%, device 6 is 100%; etc.). In various embodiments, the diagnoses 326a-326n can be stored in a data repository 354 and / or provided to various nodes 330a-330n, such as operator nodes, training supervisor nodes, data system nodes, etc. In some embodiments, the nodes 330a-330n can be operated by a training supervisor or other subject matter expert, who can review, annotate, or otherwise analyze the diagnoses 326a-326n, channel data, etc. In various embodiments, the training supervisor can label the channel data, diagnoses, etc. to improve the training data, diagnostic models, etc.

[0048] In some embodiments, one or more diagnostic units (e.g., an instrument system and / or components thereof) can generate time-based digitized binary analog data from internal devices in various possible scenarios, including but not limited to, at rest, during diagnostic testing, during normal operation (e.g., running a method), etc.

[0049] Channel data (e.g., from channel 317) can be subscribed to and read via a process "processor" (e.g., processor 324) resident on a computing device (e.g., system controller 320) that coordinates and controls multiple instruments. In some embodiments, there can be one "system controller" for each system (e.g., instrument cluster).

[0050] The processor may read one or more configuration data items to:

[0051] Training Mode – Input configuration data, specific channel data, and / or associated diagnostics stored in a data repository (local / cloud-based) into one or more neural networks to create a model for each instrument / system / device diagnostic. The training supervisor (e.g., accessing information via Node A 330a) can be a subject matter expert who helps label the initial training data. Machine learning models can be created at any time, such as at set intervals, in the background, during downtime, or in the cloud; or

[0052] Test Mode – Uses configuration data and specific real-time channel data input into one or more neural networks to predict one or more diagnoses using the models created for each instrument / system / device. The test channel data can be added to a training data repository (e.g., data repository 354) along with the diagnoses. Testing the model y = f(x) can include applying f to a new test case x to output a predicted value y.

[0053] The processed diagnostics for the entire instrument / system and each device on that instrument / system can be reported to data consumers, such as operators, connected data systems, training supervisors, the instrument / system itself, etc. As the training data repository becomes larger ("big data"), the neural network model becomes more accurate. The training data repository can be a local database or a remote database (i.e., Amazon Web Services or similar).

[0054] Figure 4 The configuration structure for a diagnostic device is shown. For example, configuration 405 may be a configuration file for an analytical instrument (such as an LC-MS instrument). In various embodiments, configuration 405 may describe properties of one or more diagnostic models, such as inputs, outputs, channels, etc. (e.g., in a *.json file). In various embodiments, a diagnostic device may be associated with multiple configurations, including but not limited to a capture configuration, a training configuration, and a training configuration.

[0055] In some embodiments, the first step in the diagnostic process according to some embodiments may be to create various models (e.g., one model for each component of the instrument or diagnostic device). Configuration 405 may be used to list, among other things, the inputs / outputs to create the model for each diagnostic device. In various embodiments, the inputs may include sensor data, instrument flow / age / alarm code history, etc. After configuration 405 is created, many records of good examples of diagnostic operation may be captured. For example, for a pump, a software routine may be run (at power-up / system startup / injection / self-organization) to flow at 1 ml / min for 1 minute. The diagnostic service application and / or training supervisor may mark all diagnostic columns as good (1) (see, e.g., Figure 7 and Figure 14). A mass flow meter can optionally be used to verify that the flow rate is ideal, ensuring that the diagnostic service application and / or training supervisor is flagging appropriate "good" examples. Many recordings of bad scenarios can then be captured for each component using the same criteria (e.g., flowing at 1 ml / min for 1 minute). A neural network can then be trained on the captured data to create a model. Finally, when a new capture is completed at the customer site (at power-up / system startup / injection / self-organization), the data can be fed into a diagnostic model (e.g., a neural network) which is run against each model to give a diagnostic metric (e.g., percentage normal) value for each component in the instrument / system.

[0056] Example: Diagnostic Scenario for a Chromatography System

[0057] Figure 5 Graph 505 shows a real-time graph of a chromatography system for various diagnostic devices 510, according to some embodiments. Figure 6 and Figure 7 Data capture files 605 and 705 are shown, respectively, related to the data plotted in chart 505. In various embodiments, the diagnosis column may have a value of 1 (if normal) and 0 (if unhealthy). The initial training dataset may have the diagnosis column data modified by the diagnostic service application and / or subject matter experts, and capture various "priming" examples of unhealthy diagnoses to help obtain optimal training data for model generation. In various embodiments, this may only be done initially, as under normal operation, the diagnosis data assignment may be automatic. Figure 8 Depicted is a diagnostic output 805 associated with a prediction associated with the information plotted in graph 505 . Figure 9 Graph 905 shows a "bad transducer" scenario (eg, transducer output is 0) according to some embodiments, and Figure 10 Data capture file 1005 associated with the data plotted in chart 905 is shown. Figure 11 A diagram 1105 illustrating a "bad encoder" scenario according to some embodiments, and Figure 12 A data capture file 1205 (eg, post-supervision) associated with the data plotted in the graph 1105 is shown.

[0058] Figure 13 Graph 1305 illustrates a "loss of prime" scenario according to some embodiments. For example, diagnosing prime may involve determining when solvent is depleted at the solvent supply. Figure 13 In the case of a high pressure, the accumulator pressure may not be maintained, resulting in fluctuations. Figure 14As shown in the data capture file 1405 , the diagnostic service application and / or the training supervisor may mark the diagnostic column “Diagnostic Device Diagnostic Perfusion” of the channel as bad (0).

[0059] Figure 15 Graph 1505 illustrates a "bad check valve" (e.g., for an LC pump) scenario according to some embodiments. For example, in Figure 13 In some scenarios, the accumulator pressure can have >1% "fluctuations". For example, a "bad check valve" diagnostic can occur when the check valve ball is not seated correctly, causing fluid to be siphoned during the pump stroke. The condition of the check valve can be correlated to its age and historical flow rates. Figure 16 As shown in the data capture file 1605 , the diagnostic service application and / or the training supervisor may mark the diagnostic column “Diagnostic Device Diagnostic Check Valve” of the channel as bad (0).

[0060] Another scenario for pump component diagnostics relative to flow for 1 minute in some embodiments is capturing training and test data during a leak test. Figure 13 and Figure 15 The same or similar channels in a pump may however employ different profiles, using different diagnostic models. The pump leak test performs a pressure ramp and then employs feedback control to maintain a constant pressure within the pump head. By monitoring the plunger travel during the constant pressure phase, it calculates the leak rates of the main and accumulator actuators. These leak rates may indicate if there is a fault in the check valve, tubing, fittings, plunger, or plunger seal. The diagnostic service application and / or training supervisor may mark the diagnostic column "Diagnostic Device Diagnostic Flow Accuracy" for that channel as bad (0).

[0061] In various embodiments, the diagnostic for a "bad flow accuracy" condition may involve a situation that prevents the full flow of solvent delivered by the pump from passing through. An example of a "bad flow accuracy" condition is capturing a one minute flow of 1 ml / min where, for example, the accumulator pressure and / or main pressure are low and / or the encoder acceleration is low. In an exemplary embodiment, a "bad seal" diagnostic may occur when the pump piston chamber has worn seals, causing fluid to be siphoned when the pump is stroked. The condition of the pump seals may be related to their age and historical flow rates. The diagnostic service application and / or training supervisor may mark the diagnostic column "Diagnostic Device Diagnostic Seals" for that channel as bad (0). A "bad column" diagnostic may occur when the column has degraded to the point where it is either over-compressed or allows flow. This is typically influenced by the age of the column and the historical flow rates through the column. Such diagnostics may include capturing a "bad column" example of a one minute flow of 1 ml / min. In one example, the accumulator pressure may have significant pressure fluctuations or be higher than normal pressure. The diagnostic service application and / or training supervisor may mark the diagnostic column "Diagnostic Device Diagnostic Column" for that channel as bad (0).

[0062] According to some embodiments, various detector instruments can be used, such as ultraviolet (UV) detectors, LC detectors that measure AU (absorbance units), and / or detectors that can be used to capture / train / test configuration files for TUV (tunable UV) FLR, PDA detectors, etc. For example, in some embodiments, each diagnostic device (e.g., instrument) can be associated with a configuration file for its diagnostic settings (e.g., data_analyzer_config_diagnostic_capture_file_tuv.json, data_analyzer_config_diagnostic_train_file_tuv.json, data_analyzer_config_diagnostic_test_file_tuv.json, etc.).

[0063] Figures 17 to 39 An illustrative diagnostic scenario for a chromatography system according to some embodiments is shown.

[0064] Figure 40 An embodiment of an exemplary computing architecture 4000 suitable for implementing various embodiments as described above is shown. In various embodiments, computing architecture 4000 may include or be implemented as part of an electronic device. In some embodiments, computing architecture 4000 may represent, for example, computing device 110. The embodiments are not limited in this context.

[0065] As used herein, the terms "system," "component," and "module" are intended to refer to a computer-related entity, which may be hardware, a combination of hardware and software, software, or software in execution, examples of which are provided by exemplary computing architecture 4000. For example, a component may be, but is not limited to, a process running on a processor, a processor, a hard drive, multiple storage drives (optical and / or magnetic storage media), an object, an executable file, a thread of execution, a program, and / or a computer. For example, both an application running on a server and the server may be a component. One or more components may reside within a process and / or thread of execution, and components may be localized on a single computer and / or distributed across two or more computers. Furthermore, components may be communicatively coupled to each other via various types of communication media to coordinate operations. Coordination may involve unidirectional or bidirectional information exchange. For example, components may communicate information in the form of signals transmitted via the communication media. This information may be implemented as signals assigned to various signal lines. In such an assignment, each message is a signal. However, alternative embodiments may employ data messages instead. Such data messages may be sent via various connections. Exemplary connections include parallel interfaces, serial interfaces, and bus interfaces.

[0066] The computing architecture 4000 includes various general-purpose computing elements, such as one or more processors, multi-core processors, co-processors, memory units, chipsets, controllers, peripherals, interfaces, oscillators, timing devices, video cards, audio cards, multimedia input / output (I / O) components, power supplies, etc. However, embodiments are not limited to the implementation of the computing architecture 4000.

[0067] like Figure 40 As shown, the computing architecture 4000 includes a processing unit 4004, a system memory 4006, and a system bus 4008. The processing unit 4004 can be any of various commercially available processors, including but not limited to: AMD ® , Athlon ® 、Duron ® and Opteron ® Processor: ARM ® Application, Embedded and Security Processors; IBM ® and Motorola ® DragonBall ® and PowerPC ® Processors; IBM and Sony ® Cell processor; Intel ® Celeron ® 、Core(2) Duo ® , Itanium ®、Pentium ® 、Xeon ® and XScale ® Processor; and similar processors. Dual microprocessors, multi-core processors, and other multi-processor architectures can also be used as the processing unit 4004.

[0068] The system bus 4008 provides an interface for system components, including but not limited to an interface connecting the system memory 4006 to the processing unit 4004. The system bus 4008 can be any of several types of bus structures, which can be further interconnected to a memory bus (with or without a memory controller), a peripheral bus, and a local bus using any of a variety of commercially available bus architectures. Interface adapters can be connected to the system bus 4008 via a slot architecture. Exemplary slot architectures can include but are not limited to Accelerated Graphics Port (AGP), Card Bus, (Extended) Industry Standard Architecture ((E)ISA), Micro Channel Architecture (MCA), NuBus, Peripheral Component Interconnect (Extended) (PCI(X)), PCI Express, Personal Computer Memory Card International Association (PCMCIA), and the like.

[0069] The system memory 4006 may include various types of computer-readable storage media in the form of one or more higher-speed memory cells, such as read-only memory (ROM), random-access memory (RAM), dynamic RAM (DRAM), double-data-rate DRAM (DDRAM), synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, polymer memory (such as ferroelectric polymer memory, bidirectional memory, phase change or ferroelectric memory), silicon-oxide-nitride-oxide-silicon (SONOS) memory, magnetic or optical cards, device arrays (such as redundant array of independent disks (RAID) drives), solid-state memory devices (e.g., USB memory, solid-state drives (SSDs), and any other type of storage medium suitable for storing information. Figure 40 In the illustrated embodiment, system memory 4006 may include non-volatile memory 4010 and / or volatile memory 4012. A basic input / output system (BIOS) may be stored in non-volatile memory 4010.

[0070] Computer 4002 may include various types of computer-readable storage media in the form of one or more relatively slow memory units, including an internal (or external) hard disk drive (HDD) 4014, a magnetic floppy disk drive (FDD) 4016 for reading from or writing to a removable magnetic disk 4018, and an optical drive 4020 (e.g., a CD-ROM or DVD) for reading from or writing to a removable optical disk 4022. HDD 4014, FDD 4016, and optical drive 4020 may be connected to system bus 4008 via HDD interface 4024, FDD interface 4026, and optical drive interface 4028, respectively. HDD interface 4024 for external drive implementations may include at least one or both of Universal Serial Bus (USB) and IEEE 1384 interface technologies.

[0071] The drives and associated computer-readable media provide volatile and / or nonvolatile storage of data, data structures, computer-executable instructions, etc. For example, a number of program modules may be stored in the drives and memory units 4010, 4012, including an operating system 4030, one or more application programs 4032, other program modules 4034, and program data 4036. In one embodiment, the one or more application programs 4032, other program modules 4034, and program data 4036 may include, for example, various applications and / or components of the computing device 110.

[0072] A user may enter commands and information into the computer 4002 through one or more wired / wireless input devices, such as a keyboard 4038 and a pointing device such as a mouse 4040. Other input devices may include a microphone, an infrared (IR) remote control, a radio frequency (RF) remote control, a game controller, a stylus, a card reader, a dongle, a fingerprint reader, gloves, a graphics tablet, a joystick, a keyboard, a retina reader, a touch screen (e.g., capacitive, resistive, etc.), a trackball, a trackpad, sensors, a stylus, and the like. These and other input devices are typically connected to the processing unit 4004 through an input device interface 4042 coupled to the system bus 4008, but may be connected through other interfaces such as a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR port, and the like.

[0073] A monitor 4044 or other type of display device is also connected to the system bus 4008 via an interface, such as a video adapter 4046. The monitor 4044 may be internal or external to the computer 4002. In addition to the monitor 4044, computers typically include other peripheral output devices, such as speakers, printers, and the like.

[0074] Computer 4002 can operate in a networked environment using logical connections via wired and / or wireless communications to one or more remote computers, such as remote computer 4048. Remote computer 4048 can be a workstation, server computer, router, personal computer, portable computer, microprocessor-based entertainment device, peer device, or other public network node, and typically includes many or all of the elements described with respect to computer 4002, but only memory / storage device 4050 is shown for simplicity. The depicted logical connections include wired / wireless connections to a local area network (LAN) 4052 and / or a larger network, such as a wide area network (WAN) 4054. Such LAN and WAN networking environments are common in offices and companies and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a global communications network, such as the Internet.

[0075] When used in a LAN networking environment, the computer 4002 is connected to the LAN 4052 through a wired and / or wireless communication network interface or adapter 4056. The adapter 4056 can facilitate wired and / or wireless communication to the LAN 4052, which can also include a wireless access point provided thereon for communicating with the wireless functionality of the adapter 4056.

[0076] When used in a WAN networking environment, the computer 4002 may include a modem 4058, or be connected to a communications server on the WAN 4054, or have other means for establishing communications over the WAN 4054, such as over the Internet. The modem 4058, which may be internal or external and a wired and / or wireless device, is connected to the system bus 4008 via the input device interface 4042. In a networked environment, program modules depicted relative to the computer 4002, or portions thereof, may be stored in the remote memory / storage device 4050. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used.

[0077] The computer 4002 is operable to communicate with wired and wireless devices or entities using the IEEE 802 family of standards, such as wireless devices operatively configured for wireless communications (e.g., IEEE 802.16 over-the-air modulation techniques). This includes at least Wi-Fi (or Wireless Fidelity), WiMax, and Bluetooth. ™Wireless technologies, such as Wi-Fi, can be predefined structures like conventional networks, or simply ad hoc communication between at least two devices. Wi-Fi networks use radio technologies known as IEEE 802.11x (a, b, g, n, etc.) to provide secure, reliable, and fast wireless connectivity. Wi-Fi networks can be used to connect computers to each other, to the internet, and to wired networks (using IEEE 802.3-related media and features).

[0078] Numerous specific details have been set forth herein to provide a thorough understanding of the embodiments. However, those skilled in the art will appreciate that the embodiments may be practiced without these specific details. In other instances, well-known operations, components, and circuits have not been described in detail to avoid obscuring the embodiments. It will be understood that the specific structural and functional details disclosed herein may be representative and do not necessarily limit the scope of the embodiments.

[0079] The expressions "coupled" and "connected," as well as their derivatives, may be used to describe some embodiments. These terms are not intended to be synonymous with each other. For example, some embodiments may be described using the terms "connected" and / or "coupled" to indicate that two or more elements are in direct physical or electrical contact with each other. However, the term "coupled" may also mean that two or more elements are not in direct contact with each other, but still cooperate or interact with each other.

[0080] Unless expressly stated otherwise, it is understood that terms such as "process," "calculate," "calculate," "determine," etc. refer to the operations and / or processes of a computer or computing system or similar electronic computing device that process and / or transform data represented as physical quantities (e.g., electronic) within the registers and / or memories of the computing system into physical quantities similarly represented within the memories, registers, or other such information storage devices, transmission, or display devices of the computing system. The embodiments are not limited in this context.

[0081] It should be noted that the methods described herein do not have to be executed in the order described or in any particular order. In addition, the various activities described with respect to the methods identified herein can be executed in series or in parallel.

[0082] Although specific embodiments have been depicted and described herein, it will be understood that any arrangement that is expected to achieve the same purpose may replace the specific embodiments shown. The present disclosure is intended to encompass any and all adaptations or variations of various embodiments. It will be understood that the above description is provided in an illustrative manner and is not restrictive. After reading the above description, the combination of the above-described embodiments and other embodiments not specifically described herein will be apparent to those skilled in the art. Therefore, the scope of the various embodiments includes any other application in which the above-described compositions, structures, and methods are used.

[0083] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts are disclosed as example forms of implementing the claims.

Claims

1. A device comprising: at least one memory; and Logic, coupled to the at least one memory, configured to: receiving diagnostic information associated with at least one analytical instrument, and processing the diagnostic information using a computational model to generate at least one diagnostic model for at least one diagnostic device; The analytical instrument includes a mass spectrometer or a chromatographic device; The computational model is a machine learning model; The diagnostic information includes standard operation information for the at least one diagnostic device and abnormal operation information for the at least one diagnostic device. 2 . The apparatus of claim 1 , wherein the logic is configured to receive at least one status indicator of the diagnostic information, the at least one status indicator being configured to indicate whether the diagnostic information is standard operation information or abnormal operation information.

3. The apparatus of claim 2, the logic to generate the at least one diagnostic model by training the computational model with the standard operating information based on the at least one status indicator to model a standard operating condition of the at least one diagnostic device.

4. The apparatus of claim 2, the logic to generate the at least one diagnostic model by training the computational model with the abnormal operation information based on the at least one status indicator to model an abnormal operating condition of the at least one diagnostic device.

5. The apparatus of claim 2, wherein the logic is configured to generate the at least one diagnostic model by training the computational model with the standard operating information and the abnormal operating information based on the at least one status indicator to model standard operating conditions and abnormal operating conditions of the at least one diagnostic device. The apparatus of claim 1 , said logic to receive operational information associated with the at least one diagnostic device. 7 . The apparatus of claim 6 , said logic to process the diagnostic information using the at least one diagnostic model to generate an operational value for the diagnostic.

8. The apparatus of claim 6, said logic to update said at least one diagnostic model using said operational information. 9 . The apparatus according to claim 7 , wherein the operation value is used to indicate whether the diagnosis is standard or abnormal.

10. The apparatus according to claim 7, wherein the operation value is used to indicate a standard percentage or an abnormal percentage of the diagnosis.

11. The apparatus of claim 1 , said logic to access configuration information for determining an input or output of the at least one diagnostic model.

12. The apparatus of claim 1, the at least one analytical instrument comprising at least one of a scientific laboratory instrument, a chromatography system, a liquid chromatography system, a gas chromatography system, a mass analyzer, a mass spectrometer, or an ion mobility spectrometer.

13. The apparatus of claim 1, the diagnostic information comprising at least one of operational information or analytical information.

14. The apparatus of claim 1, the diagnostic information comprising operational information associated with components of the at least one analytical instrument. The apparatus according to claim 14 , wherein the operation information comprises component configuration information for configuring the operation of the component. The apparatus according to claim 14 , wherein the operation information comprises component operation information associated with an operation of the component.

17. The apparatus of claim 14, said logic to receive the operational information via at least one data channel in communication with the component.

18. The apparatus of claim 14, said logic to receive the operational information via at least one sensor in communication with the component.

19. The apparatus of claim 14, said logic to receive the operational information via at least one configuration file for the component.

20. The apparatus of claim 14, said logic to receive the operational information during a test run of the at least one analytical instrument.

21. The apparatus of claim 14, said logic to receive said operational information during an analytical run of said at least one analytical instrument.

22. The device of claim 14, wherein the component comprises at least one of a pump, a fluid source, an accumulator, a valve, a check valve, a solenoid valve, a motor, an encoder, an actuator, a sensor, a power supply, a seal, a chromatography column, a detector, an electrode, a heating element, or a cooling element.

23. The apparatus of claim 16, the component operation information comprising at least one of: flow rate, pressure, temperature, fluid volume, voltage, amperage, position, or status.

24. The apparatus of claim 1, wherein the at least one diagnostic information comprises an operating characteristic of the at least one analytical instrument.

25. The apparatus of claim 1, wherein the at least one diagnostic information comprises an operating characteristic of a component of the at least one analytical instrument.

26. The apparatus of claim 1, the diagnostic information comprising analytical information comprising information associated with an analytical output generated by the at least one analytical instrument.

27. The apparatus of claim 26, the analytical output comprising at least one of a chromatogram or a mass spectrum.

28. The apparatus of claim 26, wherein the analytical information comprises at least one of peak retention time, peak area, peak shape, peak width, system pressure, or peak number.

29. The apparatus of claim 26, said logic to process the analytical information using the at least one diagnostic model to determine a fault indicator.

30. The apparatus of claim 29, the fault indicator indicating a change in a reference value associated with the analysis information.

31. The apparatus of claim 29, wherein the analytical information comprises a chromatogram or a mass spectrum, and the fault indicator indicates one of a plurality of fault indicator categories, wherein the fault indicator category comprises one or more of a change in peak retention time, a change in peak area, a change in peak shape, a change in peak width, a change in system pressure, and a change in the number of peaks.

32. The apparatus of claim 29, said logic to determine at least one cause associated with the fault indicator.

33. The apparatus of claim 29, said logic to perform a test sequence based on the failure indicator to determine at least one cause associated with the failure indicator.

34. The apparatus of claim 29, said logic to determine at least one fault fingerprint associated with the fault indicator.

35. The device of claim 1, the computational model comprising at least one of a deep learning (DL) computational model, a neural network, a convolutional neural network (CNN), a recurrent neural network (RNN), or a multilayer perceptron (MLP) model.

36. The apparatus of claim 1, said logic to present a diagnosis on a display device operatively coupled to said logic.

37. The apparatus of claim 1, the at least one diagnostic model comprising at least one input and at least one output.

38. The apparatus of claim 37, the at least one input comprising a data channel of a component of the at least one analytical instrument associated with the diagnosis.

39. A computer-implemented method comprising performing the following operations by logic of an apparatus according to any one of claims 1 to 38: receiving diagnostic information associated with at least one analytical instrument, and processing the diagnostic information using a computational model to generate at least one diagnostic model for at least one diagnostic device; The analytical instrument includes a mass spectrometer or a chromatographic device; Computational models are machine learning models.

40. A system comprising the apparatus according to any one of claims 1 to 38, the at least one analytical instrument comprising a plurality of analytical instruments.

41. The system of claim 40, the plurality of analytical instruments comprising at least one of a chromatography system and a mass analyzer system.

42. A computer-implemented method comprising performing, by a system controller of an apparatus: capturing training operational data of components of the analytical instrument; generating diagnostic data by indicating whether the training operational data is standard or abnormal; generating a diagnostic model by training a computational model using the diagnostic data; receiving real-time operational data of the component; as well as predicting at least one diagnostic information of the component by processing the real-time operational data using the diagnostic model; The analytical instrument includes a mass spectrometer or a chromatographic device; The computational model is a machine learning model; The diagnostic information includes standard operation information for at least one diagnostic device and the diagnostic information includes abnormal operation information for at least one diagnostic device.

43. An apparatus comprising: at least one memory; and Logic, coupled to the at least one memory, configured to: receiving training operational information analyzing operational characteristics of components of the instrument, receiving at least one training indicator to specify whether the training operation information is standard or abnormal for the operation characteristic, generating diagnostic information based on the operation information and the at least one training indicator, and generating a diagnostic model for the operational characteristic by training a computational model using the diagnostic information; The analytical instrument includes a mass spectrometer or a chromatographic device; The computational model is a machine learning model; The diagnostic information includes standard operation information for at least one diagnostic device and the diagnostic information includes abnormal operation information for at least one diagnostic device.

Images