Test and measurement device, method for predicting the integrity of a test and measurement device, and test and measurement system

Abstract

Predict the service needs for test and measurement equipment. [Solution] The test measurement device 10 comprises hardware components consisting of one or more system main circuit boards 16, a user interface 122, a display 20, a power supply unit 168, one or more ADCs 162, and a memory 166; one or more artificial intelligence (AI) models stored in the memory 166; and one or more processors 164 configured to receive input from at least one hardware component and execute a program that causes one or more processors 164 to perform processing that operates one or more AI models to output the predicted service needs of the test measurement device 10.

Description

Technical Field

[0001] The present disclosure relates to test and measurement equipment, and in particular to systems and methods for predicting hardware failures and other service needs of test and measurement equipment such as oscilloscopes.

Background Art

[0002] Hardware failures of test and measurement equipment such as oscilloscopes depend on various conditions and parameters. To repair customers' test and measurement equipment in a timely manner, service teams need to maintain a hardware inventory list in preparation for future repairs. If the required hardware is not available, customers have to wait a long time to get their equipment back, which may disrupt the customers' work.

[0003] Calibration is also one area where customers need to send their equipment to a laboratory for calibration. This is a regular recommendation by manufacturers of oscilloscope equipment. Depending on the model of the oscilloscope, it may be necessary to choose between laboratory or in-house calibration. Calibration generally depends on various parameters, such as the usage of the equipment, the temperature range, and the hardware performance of the equipment. For this reason, customers need to determine the need for calibration, which is unpredictable. If the accuracy of measurement results is suddenly lost, customers may have to stop using the equipment without prior planning.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Non-Patent Documents

[0005]

Non-Patent Document 1

[0006] Therefore, it is necessary to predict the service needs (demand) for testing and measurement equipment. [Means for solving the problem]

[0007] Generally, embodiments of the present disclosure operate to predict the health of internal hardware in order to detect hardware failures of test and measurement devices at an early stage. As used in the present disclosure, the prediction of “service needs” relating to a test and measurement device includes both the prediction of the need for repairs, including preventative repairs of components, and the need for calibration of the device or its components. The term “service needs” as used in this application encompasses all of these aspects of the test and measurement device and its components.

[0008] Embodiments of the technology disclosed herein generally enable the detection of calibration needs for equipment with high accuracy. An artificial intelligence model, referred to herein as the “Model” or “AI Model,” predicts calibration health as a percentage. If calibration health falls below a certain percentage, this indicates that the system requires calibration at the customer’s site or service depot (such as the laboratory or other facility of the test measurement equipment supplier).

[0009] Furthermore, some embodiments of this disclosure may include a model in which data can be acquired from the health monitoring system of each device and stored in a central server system. This system includes not only the location information of the devices but also health and calibration prediction information for each device on the market. A service team can use this system to assess the health of the internal hardware of various devices on the market. It is also possible to segment the data based on region.

[0010] For example, suppose a customer is currently using a 2034 mixed-signal oscilloscope (MSO). By integrating the health forecast reports for each oscilloscope, the service team can determine that the power supply units (PSUs) of 100 mainboards or carrier boards are below 70%. This indicates that the service team anticipates that 100 MSO oscilloscopes in a particular area will require repair due to a specific hardware failure. The same applies to the calibration of test and measurement equipment. The system in this embodiment can assist the service team in scheduling and procuring replacement parts and planning the appropriate undertaking capacity of test and calibration stations.

[0011] Embodiments disclosed herein construct one or more AI models based on this data. These neural network-based models are useful for predicting future needs and demands.

[0012] This will be helpful in making important decisions regarding future product needs related to new development areas at the factory. For example, it will be possible to understand the current inventory of a particular piece of equipment in the market, along with its health condition. This information can then be used to forecast future demand for these pieces of equipment in the market.

[0013] This approach focuses on improving the efficiency of specific hardware and is beneficial in providing the capability to address hardware issues and formulate strategies for future improvements. As a result, device manufacturers can produce devices with higher reliability, stability, and longer service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] [Figure 1] Figure 1 shows an embodiment of a test measurement apparatus. [Figure 2] Figure 2 is a graph showing apparatus usage versus failure probability as a function of apparatus age. [Figure 3] Figure 3 is a graph showing apparatus noise versus system age for an oscilloscope. [Figure 4] Figure 4 shows an embodiment of input parameters to a neural network in an embodiment of a model used for predicting hardware health. [Figure 5A] Figure 5A shows the trend of training error. [Figure 5B] Figure 5B shows the trend of training error. [Figure 6A] Figure 6A shows a prediction obtained using a training data set corresponding to the training error trend of Figure 5A. [Figure 6B] Figure 6B shows a prediction obtained using a training data set corresponding to the training error trend of Figure 5B. [Figure 7A] Figure 7A shows a prediction based on test data. [Figure 7B] Figure 7B shows a prediction based on test data. [Figure 8] Figure 8 is a graph showing prediction efficiency using training data and test data. [Figure 9] Figure 9 shows a display of an embodiment of a user interface for a health monitoring system. [Figure 10] Figure 10 shows another display of an embodiment of a user interface for a health monitoring system. [Figure 11]FIG. 11 shows an interface of a predicted output of internal hardware health and a predicted output of calibration health. DESCRIPTION OF EMBODIMENTS

[0015] In the following description, some examples in the health prediction function unit of the mixed-signal oscilloscope (MSO) series from Tektronix are given. However, embodiments of the presently disclosed technology are not limited to this example, which is provided for ease of understanding. The approach in the embodiments can be applied to all types of devices.

[0016] Figure 1 shows a schematic diagram of the MSO 10, but many of the boards, individual components, and processors should be understood to represent the same components in any device. The MSO has multiple system boards, generally printed circuit boards (PCBs), which consist of any type of board on which the processor and integrated circuit chips reside, and may consist of only one board. The MSO has a front panel 12, which has a user interface 122 with an interface 120 for probes used in testing, and user operating devices including knobs, buttons, switches, keypads, and other operating devices that allow the user to input information and make selections. The display 20 may consist of a touchscreen and may incorporate some or all of the user operating devices. The hardware interface 124 includes an integrated circuit that provides correction signals for the probe input signals, as well as interfaces for other hardware such as USB ports and accessory video ports. The video interface 128 consists of the necessary components for displaying input data on the display 20. Communication port 22 provides a communication or network interface, allowing the device to set up communication connections and network-based communications, including short-range wireless networks, Wi-Fi networks, and cellular networks. This enables the device to send health check data and the central service server to request health check data, details of which will be described later. The various components of each of these subsystems may be connected to each other directly or through other components not shown.

[0017] The front end 14 may include components 140 that operate in response to input signals from the probe and the device under test (DUT). These may include, but are not limited to, attenuators, preamplifiers, and amplifiers. The front end may also include a controller (processor 142) that manages the inputs and outputs of the MSO.

[0018] The main board 16 may include acquisition hardware components 160 such as a digitizer, and an analog-to-digital converter (ADC) 162 for converting analog signals into digital signals for display and analysis. The acquisition hardware may also include trigger circuits and signal conditioning circuits for filtering signals. The main board also includes one or more processors 164 and memory 166. Many other boards may also contain various types of memory, but for simplicity, memory 166 represents all the memory in the device 10. Memory 166 can store one or more artificial intelligence (AI) models.

[0019] The carrier board 18 typically includes a power supply unit (PSU) 180 that supplies power to other boards. The main board 16 may include a power supply unit (PSU) circuit 168 for converting input power, such as 12V DC, to various voltage levels as needed. The main board, and possibly the carrier board, may include one or more processors 164 and 182. One or more processors are configured to execute programs (code) that cause one or more processors within the device to perform various tasks and functions, including running and training AI models.

[0020] Table 1 is a list summarizing the models of the internal hardware of the MSO series oscilloscopes. Furthermore, in some embodiments, a known built-in signal source for "probe compensation function" may be used. According to embodiments of the present invention, AI-based models with various input parameters can predict specific hardware failures early. Timely prediction of hardware failures is beneficial for customers in planning work using the equipment. This embodiment evaluates the health and calibration of the internal hardware board. Internal Hardware List of MSO Series Oscilloscopes (Table 1) [Table 1]

[0021] In an example of an embodiment of this disclosure, a neural network (NN) model based on a list of input parameters is used, such as the example shown below. This model is trained to monitor the health of individual hardware components and predict potential failures. List of parameters (Tables 2-1 and 2-2) [Table 2-1] [Table 2-2]

[0022] Please note that the list of input parameters in the table above is merely an example. In other embodiments, the input parameters may be reduced or increased to include parameters such as fan speed, fan voltage, power supply output voltage, and temperature stress duration. The selection of several models, as detailed below, may depend on the input parameters available for a particular device.

[0023] The importance of each parameter used as input is as follows:

[0024] The SPC (Signal Path Compensation) status helps to ensure accuracy for other measurement parameters. For example, even if you measure the oscilloscope noise of a de-skew source parameter, if the SPC status is red (RED), you may choose not to use that measured parameter as input to a neural network.

[0025] The temperature of a custom ASIC is typically published in the self-test log. This may also apply to the internal temperature of a custom ASIC, such as a custom ASIC ADC. At high temperatures, the deviation of the measured value from the expected value becomes large. The same applies to the temperature of fan controller diodes.

[0026] System usage and age are used to train a neural network on how hardware ages over time. As electronic hardware ages, its components degrade. Hardware reliability decreases with usage and age. Figure 2 shows a graph illustrating the relationship between device usage, device age, and the likelihood of hardware failure. This embodiment measures how long a particular device will be used. As device usage increases, the likelihood of hardware failure also increases. Furthermore, frequent calibration may be necessary for the system to function optimally and provide accurate information to the user.

[0027] The model in this embodiment learns and trains the model with an exponential decay function. This function takes into account the probability of the device changing over time as a parameter of the model. Figure 3 shows an example of an exponential model that shows the relationship between the age of the device and the noise that appears in the device.

[0028] Many test and measurement devices, particularly MSOs, are equipped with self-service testing capabilities that users can perform at customer sites or service depots. Self-test results show a high correlation with the health and performance of individual hardware components. Self-test data is also used for model training.

[0029] MSO series oscilloscopes have a built-in "Probe Compensation" function, which generates a square wave signal with a fast edge of 1 kHz at a voltage level of + / - 1.25 volts. This can be used as a known signal source to measure the input parameters of each channel, as shown in index 4 through 8 in the table above. This function is unique to MSO, but other instruments also have a function to generate a signal from a known signal source to correct or calibrate the instrument.

[0030] By using signals from known signal sources, the operating characteristics of the device, such as voltage, fundamental frequency, and rise time / rise time, can be measured. This allows users and models to make judgments about the device's operation based on the results of these parameters. For example, if the measured voltage (minimum, maximum), fundamental frequency, or rise time / rise time deviates from expected values, it may indicate one of the following possibilities: the device may need calibration, the acquisition (waveform data acquisition) hardware within the device may be faulty, or the power distribution mechanism within the device may be faulty. The device's built-in power distribution mechanism also supplies power to the "probe compensation function." If the power supply fails to provide power to this component, it may output an incorrect voltage. Furthermore, a problem with the power distribution mechanism may cause jitter in the "differential high-speed edge" signal source. Additionally, there may be a problem with the acquisition (ACQ) function of the front channel, which may cause problems with the probe interface.

[0031] Oscilloscope noise RMS (RMS value) is measured when no signal is connected to the oscilloscope. If the measured oscilloscope noise RMS value is high, for example, as shown in the values ​​for Ch1 and Ch4 below, it indicates one or more of the following possibilities: There may be a problem with the front panel hardware, such as the acquisition board for acquiring specific channels. There may be a problem with the SPC path calibration. • There may be a problem with the power supply unit (PSU) of the main or carrier board. [Table 3]

[0032] Embodiments of this invention develop neural network models for various types of devices. In one particular example, the neural network has an architecture consisting of 25 input layers, 12 hidden layers, and 1 output layer. The output layer outputs its values. The activation function that introduces nonlinearity is the tanh (hyperbolic tangent) function, and no activation function is applied to the output layer. Other activation functions may also be used. In this embodiment, the least mean square error (LMSE) is used as the loss function.

[0033] Figure 4 shows an overall system embodiment used for training and running a model to determine the service needs of a specific device.

[0034] This model is trained on a dataset of simulation samples 30. In a particular embodiment, 8000 simulation samples are generated, of which 40% are selected as the training dataset 32, and the remainder are set aside as the test dataset 34 for testing the model. The training process 36 is repeated until the results obtained on the test dataset 34 reach a sufficiently high probability, thereby obtaining a trained neural network 38. In operation, the input parameters 40 are normalized in 42, and then the neural network 38 operates on these inputs to provide health predictions in 44. As described above, health predictions include whether there is a service need or not. Service needs include a calibration status indicating that the device needs calibration, and indications that there are faulty components (such as boards, chips, or processors) or components that are very likely to fail.

[0035] Figures 5A and 5B show the learning error trends over iterative loops for various components (in this case, carrier processor 50, carrier PSU 51, carrier memory 52, mainboard acquisition (ACQ) 53, mainboard PSU 54, front end 55, front panel 56, and calibration 57). This model achieves the version with the lowest prediction error after 200 iterations. Figures 6A-6B show predictions using the same training sets corresponding to the same elements. Figures 7A-7B show predictions for the same training sets corresponding to the same elements.

[0036] In Figures 6A-6B and 7A-7B, it can be seen that some of the prediction circles are mainly visible only at the edges of the prediction lines, i.e., in the darker gray areas. This indicates that the model's predictions are in very close agreement with the actual data. Figure 8 shows the efficiency graphs for the training and test datasets.

[0037] According to some embodiments, the above advanced model may be integrated into a user interface that operates on the device, such as the user interface examples shown in Figures 9 to 11. Figure 9 shows an embodiment of the user interface for the Health Check Options tab. Figure 9 shows that a test has been performed on the SPC parameter, and the status indicates that the SPC parameter has passed. The user can run the test using the RUN button 60, and the test status window 62 displays the test status of each component as various components have been tested.

[0038] Once the health check test is running, the user is directed to the user interface shown in Figure 10, which focuses on the model. The user can update the model in 64 and provide a path to the training data or browse the training data in 66. The test status window 68 in this interface displays the error trend. The user selects a model in this interface.

[0039] In Figure 11, the user interface allows the user to reference the model's location and provide the path for input data. The user then runs a predictive function to display predicted service needs for this device. This allows the service center to potentially adjust its operations based on future service needs.

[0040] The following table shows an example of data returned to a central database system from multiple individual health / service prediction tools operating globally across multiple test measurement devices. [Table 4]

[0041] Please note that the data requested from customer devices is solely related to device health. This tool does not require customer-specific data to be transferred to the central database. Some device-specific data, such as device name and regional location, is not unique to the customer. Data centralization is also an option. Users can run the health prediction function at their own location and track the health of their devices without sending information to a central service depot database or other repositories.

[0042] As described above, embodiments of the disclosed technology can provide one or more of the following features and advantages.

[0043] The health prediction function provides a stepwise forecast of performance and health deterioration. This function allows users to predict problems and effectively manage the equipment.

[0044] When performing self-tests on MSO ASICs (Acquisition, Signal, Memory), a failure indication typically means that the component is not functioning properly. However, the health prediction function of this embodiment goes beyond this binary choice. It compares health indicators using currently and historically measured parameters. For example, even if no problems are found in a custom acquisition ASIC, the health prediction model assesses overall health by considering changes in parameters such as oscilloscope noise, temperature deviation, and system age.

[0045] Another feature of the health prediction function is its ability to assess the health of the calibration. By analyzing the direction of deviations in various measured parameters, the prediction function helps users identify calibration needs in advance. When used locally in the instrument, this prediction function may allow users to implement a calibration process that is appropriate for them before determining whether service center-level calibration is required.

[0046] Self-testing does not detect problems with the power distribution system of the carrier interface board or the main board, but unlike self-testing, the learned health prediction function can predict the health of the power distribution system on the MSO oscilloscope.

[0047] The health prediction feature leverages the expertise of the manufacturer's hardware and service specialists during model training.

[0048] The presented neural network (NN) solution models the nonlinear behavior of the health of individual hardware components within the device. It considers both dependent and independent parameters. Furthermore, the health prediction function also takes into account temperature fluctuations in the ASIC to predict the health of the front-end and mainboard. Unlike self-testing, the predictive model provides insights into the direction of system performance that self-testing cannot offer. The model undergoes exponential learning to account for the aging and performance contrast of electronic hardware. Because the system can detect the need for calibration in advance, customers can plan their activities proactively. The central monitoring system assists the service team in planning future work. This model allows Tektronix's service team to anticipate future hardware board repair needs early on. As a result, the service team can proactively manage hardware resources.

[0049] The proposed solution also predicts the health of individual hardware components, updates them to a central server, and ensures that each region can access the data. Furthermore, this model can measure overall quality and performance.

[0050] Embodiments of the disclosed technology can operate on a specially programmed general-purpose computer, including specially created hardware, firmware, digital signal processors, or processors that operate according to programmed instructions. The terms “controller” or “processor” in this application mean microprocessors, microcomputers, ASICs, and dedicated hardware controllers, etc. Embodiments of the disclosed technology can be implemented by one or more computers (including monitoring modules) or other devices, using computer-readable data such as program modules and computer-executable instructions. Generally, program modules include routines, programs, objects, components, data structures, etc., which, when executed by a processor in a computer or other device, perform specific tasks or implement specific abstract data type expressions. Computer-executable instructions may be stored on computer-readable storage media such as hard disks, optical disks, removable storage media, solid-state memory, and RAM. As will be understood by those skilled in the art, the functions of the program modules may be combined or distributed as needed in various embodiments. Furthermore, these functions can be embodied in whole or in part in firmware or hardware equivalents such as integrated circuits or field-programmable gate arrays (FPGAs). One or more aspects of the disclosed technology can be more effectively implemented using specific data structures, such data structures are considered to be within the scope of computer-executable instructions and computer-usable data described herein.

[0051] The disclosed embodiments may, in some cases, be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored in one or more computer-readable media that can be read and executed by one or more processors. Such instructions may be referred to as computer program products. The computer-readable media described herein means any medium accessible by a computing device. For example, but not limited to, computer-readable media may include computer storage media and communication media.

[0052] Computer storage media means any medium that can be used to store computer-readable information. Examples of computer storage media include, but are not limited to, random-access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory and other memory technologies, compact disc read-only memory (CD-ROM), DVD (Digital Video Disc) and other optical disc storage devices, magnetic cassettes, magnetic tapes, magnetic disk storage devices and other magnetic storage devices, and any other volatile or non-volatile removable or non-removable media implemented by any technology. Computer storage media exclude signals themselves and temporary forms of signal transmission.

[0053] A communication medium means any medium that can be used to transmit computer-readable information. Examples of communication mediums, though not limited to them, include coaxial cables, fiber optic cables, air, or any other medium suitable for transmitting electrical, optical, radio frequency (RF), infrared, sound, or other types of signals. Examples

[0054] The following examples are provided that are useful for understanding the technology disclosed herein. These embodiments may include one or more of the examples described below, or any combination thereof.

[0055] Example 1 is a test and measurement device, One or more system circuit boards, User interface and, The display and Power supply and One or more integrated circuit chips and Hardware components having one or more of the following, Memory and One or more artificial intelligence (AI) models stored in the memory, One or more processors, which are configured to receive input from at least one of the hardware components and execute a program that causes the one or more processors to perform a process that operates one or more of the one or more AI models in order to provide an output of predictive service needs relating to the test measurement device. It is equipped with.

[0056] Embodiment 2 is the test and measurement apparatus of Embodiment 1, further comprising a network interface that enables the test and measurement apparatus to establish a communication connection with a server at a service center and transmit service needs data to the server.

[0057] Example 3 is a test and measurement apparatus of Example 1 or 2, wherein the program that causes one or more processors to perform the process of providing the output of the predicted service needs includes a program that causes one or more processors to perform the process of providing the output as the internal hardware health of the test and measurement apparatus.

[0058] Example 4 is a test and measurement apparatus according to any of Examples 1 to 3, wherein the program that causes one or more processors to perform a process that provides the output of the predicted service needs includes a program that causes one or more processors to perform a process that provides the output as the calibration status of the test and measurement apparatus.

[0059] Embodiment 5 is a test and measurement apparatus according to any of Embodiments 1 to 4, wherein one or more processors are further configured to execute a program that causes one or more processors to display a user interface on the display that has a health check option, which allows the user to have one or more processors perform a self-test of the apparatus that generates the inputs that the AI ​​model receives from the hardware components.

[0060] Example 6 is the test and measurement apparatus of Example 5, wherein the program that causes one or more processors to perform a self-test of the apparatus includes a program that causes one or more processors to perform a process of generating a signal from a known signal source and applying it to the hardware component.

[0061] Example 7 is a test measurement apparatus of Example 5, wherein one or more processors are further configured to execute a program that causes the one or more processors to perform a process that provides a user interface that allows a user to select an AI model, and a process that operates the selected AI model.

[0062] Example 8 is a test measurement apparatus according to any of Examples 1 to 7, wherein one or more processors are further configured to execute a program that causes one or more processors to perform a process to train one or more AI models.

[0063] Example 9 is a test measurement apparatus of Example 8, wherein the program that causes the one or more processors to perform a process to train the one or more AI models includes a program that causes the one or more processors to perform a process to train the one or more AI models with respect to one or more of the self-test data and past service history data.

[0064] Example 10 is a method for predicting the integrity of a test measurement device, Processing to receive input from at least one hardware component within the above-mentioned test measurement device, A process that receives input from at least one of the above hardware components and operates one or more artificial intelligence (AI) models to provide predictive service needs output for the above test measurement device, Using the above output, a process is performed to determine whether the above test and measurement device requires service, according to the above output. It is equipped with.

[0065] Example 11 is the method of Example 10, further comprising the process of establishing a communication connection with a server at a service center and the process of transmitting service needs data to the server through the network interface of the test measurement device.

[0066] Example 12 is the method of Example 10 or 11, wherein the process of operating one or more AI models to provide the output of the predictive service needs includes the process of operating one or more AI models to provide the output as the health of the internal hardware of the test measurement device.

[0067] Example 13 is a method of any of Examples 10 to 12, wherein the process of operating one or more AI models to provide the output of the predictive service needs includes the process of operating one or more AI models to provide the output as the calibration status of the test measurement device.

[0068] Example 14 is a method of any of Examples 10 to 13, further comprising the process of displaying a user interface on a display that has a health check option, which allows the user to initiate a device self-test that causes the AI ​​model to generate the inputs received by the at least one hardware component.

[0069] Example 15 is the method of Example 14, further comprising the process of generating a signal from a known signal source and applying it to the hardware component.

[0070] Example 16 is the method of Example 14, further comprising a process for providing a user interface that allows the user to select an AI model to choose, and a process for operating the selected AI model using the above input.

[0071] Implementation 17 is one of the methods from Examples 10 to 16, further comprising a process for training one or more of the above-mentioned AI models.

[0072] Example 18 is the method of Example 17, wherein the process of training one or more AI models includes the process of training one or more AI models with one or more of the self-test data and past service history data.

[0073] Example 19 is a test measurement system, One or more test and measurement devices, One or more service centers having one or more servers that communicate with one or more of the above-mentioned test and measurement devices and Equipped with, One or more of the above test and measurement devices, Hardware components and, User interface and, The display and Power supply and One or more integrated circuit chips, Communication interface, memory and One or more artificial intelligence (AI) models stored in the memory, One or more processors, which are configured to receive input from at least one of the hardware components and execute a program that causes the one or more processors to perform a process that operates one or more of the one or more AI models in order to provide an output of predictive service needs relating to the one or more test measurement devices. It has, The above server, Processing to receive predictive service needs data and device-specific data from one or more of the above-mentioned test and measurement devices, The process of monitoring the above-mentioned predictive service needs data of one or more of the above-mentioned test measurement devices, Based on the above device-specific data, a process is performed to adjust the operation of one or more of the above service centers based on the above predicted service needs. It was configured to perform the following actions.

[0074] Example 20 is a test and measurement system of Example 19, wherein the server houses a repository of AI models, and the server is configured to perform a process of updating the one or more AI models located in the memory of the one or more test and measurement devices.

[0075] All functions disclosed in the specification, claims, abstract and drawings, and all steps in any method or process disclosed, may be combined in any combination, except where at least some of such functions or steps are mutually exclusive. Each of the functions disclosed in the specification, abstract, claims and drawings may be replaced by an alternative function that serves the same, equivalent or similar purpose, unless otherwise specified.

[0076] In addition, the description in this application refers to specific features. The technologies disclosed herein should be understood to include all possible combinations of these specific features. For example, if a particular feature is disclosed in relation to a particular form, that feature may also be available in relation to other forms, as far as possible.

[0077] Furthermore, when this application refers to a method having two or more defined steps or processes, these defined steps or processes may be performed in any order or simultaneously, as long as the circumstances do not rule out such possibilities.

[0078] For the sake of explanation, specific embodiments of the disclosed technology have been illustrated and described, but it should be understood that various modifications are possible without deviating from the gist and scope of this disclosure. Therefore, the disclosed technology should not be limited to anything other than the attached claims. [Explanation of Symbols]

[0079] 10. Mixed-Signal Oscilloscope (MSO) 12 Front Panel 14 Front End 16 Mainboard 18 Carrier Boards 20 displays 22 communication ports 30 Simulation Sample Datasets 32 training datasets 34 Test Data Sets 36 Learning Process 38 Neural Networks 40 Input Parameters 42 Normalization Process 44 Health Prediction 50 Carrier Processors 51 Carrier PSU 52 Carrier Memory 53 Mainboard Acquisition Section 54 Mainboard Power Supply Unit 55 Front End 56 Front Panel 57 Calibration 120 Probe Interface 122 User Interface 124 Hardware Interfaces 128 Video Interfaces 140 Input Hardware 142 Controllers (Processors) 160 Acquisition Hardware Components 162 Analog-to-Digital Converters (ADCs) 164 One or more processors 166 memory 168 Power Supply Unit (PSU) Circuit 180 Power Supply Units (PSUs) 182 One or more processors

Claims

1. A test and measurement device, One or more system circuit boards, User interface and, The display and Power supply and One or more integrated circuit chips and A hardware component having one or more of the following, Memory and One or more artificial intelligence (AI) models stored in the memory, One or more processors, which are configured to receive input from at least one of the hardware components and execute a program that causes the one or more processors to perform a process that operates one or more of the one or more AI models in order to provide an output of predictive service needs relating to the test measurement device. A test and measurement device equipped with [the following features].

2. The test and measurement apparatus according to claim 1, further comprising a network interface that enables the test and measurement apparatus to establish a communication connection with a server of a service center and transmit service needs data to the server.

3. The test and measurement apparatus according to claim 1, wherein the program that causes one or more processors to perform the process of providing the above-mentioned output of the above-mentioned predictive service needs includes a program that causes one or more processors to perform the process of providing the above-mentioned output as the internal hardware health or calibration status of the test and measurement apparatus.

4. The test and measurement apparatus according to claim 1, wherein one or more processors are further configured to execute a program that causes one or more processors to display on the display a user interface having a health check option that allows a user to have one or more processors perform a self-test of the apparatus causing the hardware components to generate inputs that the AI ​​model receives.

5. One or more of the above processors further, A process that provides a user interface that allows the user to select an AI model, The process of running the selected AI model mentioned above. The test and measurement apparatus according to claim 4, configured to execute a program that causes one or more of the above-mentioned processors to perform the above-mentioned task.

6. The test measurement apparatus according to claim 1, wherein one or more processors are further configured to execute a program that causes one or more processors to perform a process of training one or more AI models with one or more of the self-test data and past service history data.

7. A method for predicting the integrity of a test and measurement device, Processing to receive input from at least one hardware component within the above-mentioned test measurement device, A process that operates one or more artificial intelligence (AI) models to receive input from at least one of the above hardware components and provide an output of predictive service needs related to the above test measurement device, Using the above output, a process is performed to determine whether the above test and measurement device requires service, according to the above output. A method for predicting the soundness of a test and measurement device equipped with [a specific feature / ability].

8. A method for predicting the health of a test and measurement device according to claim 7, wherein the process of operating one or more of the above-mentioned AI models to provide the output of the above-mentioned predictive service needs includes the process of operating one or more of the above-mentioned AI models to provide the output as the health or calibration status of the internal hardware of the test and measurement device.

9. One or more test and measurement devices, One or more service centers having one or more servers that communicate with one or more of the above-mentioned test and measurement devices A test and measurement system equipped with, One or more of the above-mentioned test and measurement devices, Hardware components and, User interface and, The display and Power supply and One or more integrated circuit chips, Communication interface, memory and One or more artificial intelligence (AI) models stored in the memory, One or more processors, which are configured to receive input from at least one of the hardware components and execute a program that causes the one or more processors to perform a process that operates one or more of the one or more AI models in order to provide an output of predictive service needs relating to the one or more test measurement devices. It has, The above server, Processing to receive predictive service needs data and device-specific data from one or more of the above-mentioned test and measurement devices, A process for monitoring the above-mentioned predictive service needs data of one or more of the above-mentioned test measurement devices, Based on the above device-specific data, a process is performed to adjust the operation of one or more of the above service centers based on the above predicted service needs. A test and measurement system configured to perform the following.

10. The test measurement system according to claim 9, wherein the server houses a repository of AI models, and the server is configured to perform a process of updating the one or more AI models located in the memory of the one or more test measurement devices.

Images