Inductive sensor testing system and method

Abstract

An embodiment of the present specification provides an inductive sensor testing system and method, which includes a light shielding box, a communication component and a processing component; an adjustable light source, at least one group of standard sensors and a slide rail device are provided in the light shielding box; the adjustable light source, at least one group of standard sensors and at least one group of sensors to be tested move along the slide rail device; the processing component is configured to: in response to the adjustable light source being started according to target parameters, obtain a first reading of at least one group of standard sensors and a second reading of at least one group of sensors to be tested; the target parameters include the light intensity of the adjustable light source; based on the first reading and the second reading, determine the reading accuracy of at least one group of sensors to be tested; in response to at least one reading accuracy meeting the warning condition, issue a prompt message; wherein, the difference between the number of standard sensors and the number of sensors to be tested is greater than a preset value, and the preset value is determined based on the light intensity.

Description

Technical Field

[0001] This specification relates to the field of sensor testing, and in particular to an inductive sensor testing system and method. Background Art

[0002] As essential components in various industries and daily life, inductive sensors' precise and stable performance plays a decisive role in their applications across numerous fields. In recent years, with the rapid development of intelligent and automated technologies, the application scope of inductive sensors has become increasingly broad, and the requirements for their performance have become increasingly stringent. Therefore, efficient and accurate testing of inductive sensors is particularly important.

[0003] During the testing of inductive sensors, one of the key points is to compare the readings of the target sensor to be tested with the standard readings. However, the acquisition of readings is often performed separately. During this process, due to the instability of the light source, the light emitted may be different, resulting in test errors. Multiple tests and separate readings may lead to low test efficiency and reduced accuracy.

[0004] Therefore, it is hoped to provide an inductive sensor testing system and method that can improve testing efficiency and ensure testing accuracy. Summary of the Invention

[0005] One embodiment of the present specification provides an inductive sensor testing system, comprising a light shielding box, a communication component, and a processing component; the light shielding box is provided with an adjustable light source, at least one group of standard sensors, and a slide rail device, and at least one group of sensors to be tested is placed in the light shielding box for testing; the adjustable light source emits non-parallel light into the light shielding box; the adjustable light source, the at least one group of standard sensors, and the at least one group of sensors to be tested move along the slide rail device; the processing component is communicatively connected to the adjustable light source, the at least one group of standard sensors, and the at least one group of sensors to be tested via the communication component; the processing component is configured to: in response to the adjustable light source being activated according to a target parameter, obtain a first reading of the at least one group of standard sensors and a second reading of the at least one group of sensors to be tested; the target parameter includes the light intensity of the adjustable light source; determine the reading accuracy of the at least one group of sensors to be tested based on the first reading and the second reading; and issue a prompt message in response to at least one of the reading accuracy meeting a warning condition; wherein the difference between the number of the standard sensors and the number of the sensors to be tested is greater than a preset value, and the preset value is determined based on the light intensity.

[0006] One of the embodiments of this specification provides a sensing sensor testing method, which is executed by a processing component of the sensing sensor, including: in response to an adjustable light source being activated according to target parameters, obtaining first readings of at least one group of standard sensors and second readings of at least one group of sensors to be tested; the target parameters include the light intensity of the adjustable light source; based on the first readings and the second readings, determining the reading accuracy of the at least one group of sensors to be tested; in response to at least one of the readings meeting a warning condition, issuing a prompt message; wherein the difference between the number of the standard sensors and the number of the sensors to be tested is greater than a preset value, and the preset value is determined based on the light intensity. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein:

[0008] Figure 1 is an exemplary module diagram of an inductive sensor testing system according to some embodiments of this specification;

[0009] Figure 2 is an exemplary flow chart of an inductive sensor testing method according to some embodiments of this specification;

[0010] Figure 3 is a schematic diagram of determining setting data of a standard sensor according to some embodiments of this specification;

[0011] Figure 4 This is a schematic diagram of determining the reading accuracy of a sensor to be tested according to some embodiments of this specification. DETAILED DESCRIPTION

[0012] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0013] It should be understood that the terms "system," "device," "unit," and / or "module" used herein are a method for distinguishing different components, elements, parts, portions, or assemblies at different levels. However, if other terms can achieve the same purpose, the terms may be replaced by other expressions.

[0014] As used in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not refer to the singular but also include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0015] Flowcharts are used throughout this specification to illustrate the operations performed by systems according to embodiments of this specification. It should be understood that preceding or following operations do not necessarily need to be performed in exact order. Instead, the steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0016] Figure 1 1 is an exemplary module diagram of an inductive sensor testing system according to some embodiments of this specification. Figure 1 As shown, the inductive sensor testing system 100 may include a light shielding box 110 , a communication component 120 , and a processing component 130 .

[0017] Light shielding box 110 is a device used to provide a stable environment for experiments. Light shielding box 110 can be enclosed and contains an adjustable light source 111, at least one standard sensor 112, and at least one sensor to be tested 114. The at least one sensor to be tested 114 is placed in light shielding box 110 for testing.

[0018] The adjustable light source 111 is a component for emitting test light. In some embodiments, the adjustable light source 111 emits non-parallel light into the light shielding box 110. In some embodiments, the adjustable light source 111 emits light based on target parameters.

[0019] For more information about the parameters of the light source to be measured, please refer to Figure 3 and its related descriptions.

[0020] Standard sensor 112 is a sensor of the same type as sensor under test 114 that has a standard reading. In some embodiments, standard sensor 112 is used to provide reference data. The accuracy of test data (e.g., a second reading) of sensor under test 114 is determined based on the standard data (e.g., a first reading) of standard sensor 112.

[0021] More information about the sensors under test, reading accuracy, etc. can be found in Figure 2 and its related descriptions.

[0022] The slide rail assembly 113 guides the movement of the sensors. In some embodiments, the adjustable light source 111, at least one standard sensor 112, and at least one sensor under test 114 move along the slide rail assembly. In some embodiments, the slide rail assembly 113 can be used to adjust the positions of the adjustable light source 111, the standard sensor 112, and the sensor under test 114. The position of the sensor under test 114 can be determined based on the parameters of the light source under test and the test requirements.

[0023] For more information about the parameters of the light source to be tested and the test requirements, please refer to Figure 3 and its related descriptions.

[0024] The communication component 120 is a component that realizes the communication connection between various components. In some embodiments, the various components may include a processing component 130, a standard sensor 112, an adjustable light source 111, etc.

[0025] In some embodiments, the communication component 120 includes any suitable network, terminal device, etc. that can facilitate information and / or data exchange with the inductive sensor testing system 100. In some embodiments, one or more components of the inductive sensor testing system 100 (e.g., an adjustable light source, a standard sensor, a sensor under test, etc.) can exchange information and / or data with the processing component 130 of the inductive sensor testing system 100 via the communication component 120.

[0026] The processing component 130 refers to a component with computing capabilities, such as a computer, a computing cloud platform, software, etc. The processing component 130 may include a processing device, such as a CPU, etc. The processing component 130 is used to execute the inductive sensor testing method and process information and / or data related to the inductive sensor testing system.

[0027] In some embodiments, the processing component 130 can process data, information, and / or processing results obtained from other devices or system components and execute program instructions based on the data, information, and / or processing results to perform one or more functions described in this specification. For example, the processing component 130 can obtain a first reading of at least one standard sensor and a second reading of at least one sensor under test; determine the accuracy of the reading of at least one sensor under test; and issue a prompt message.

[0028] In some embodiments, the processing component 130 can obtain a first reading of at least one standard sensor 112 and a second reading of at least one sensor to be tested 114 in response to the adjustable light source 111 being started according to the target parameters, determine the reading accuracy of at least one sensor to be tested 114 based on the first reading and the second reading, and issue a prompt message in response to at least one reading accuracy meeting the warning condition, wherein the difference between the number of standard sensors and the number of sensors to be tested is greater than a preset value, and the preset value is determined based on the light intensity.

[0029] In some embodiments, the processing component 130 can determine the setting data of at least one group of standard sensors by referring to a database based on the target parameters and the position distribution data of at least one group of sensors to be tested in the light shielding box, and the setting data represents the position distribution of at least one group of standard sensors in the light shielding box.

[0030] In some embodiments, the processing component 130 can also grid the light shielding box based on the light shielding parameters of the light shielding box to obtain at least one grid, and then evaluate the credibility of the at least one grid based on the target parameters and setting data, and then determine at least one second reference vector based on the light loss of the at least one grid, the number of sensors to be tested and the credibility.

[0031] In some embodiments, the processing component 130 can determine the theoretical reading of the sensor to be tested based on the first reading, setting data of at least one set of standard sensors, position distribution data of at least one sensor to be tested, and target parameters, and then determine the reading accuracy based on the theoretical reading and the second reading.

[0032] In some embodiments, the processing component 130 can also construct a prediction map based on the first reading, setting data of at least one group of standard sensors, position distribution data of at least one group of sensors to be tested, and target parameters, and then determine the theoretical reading based on the prediction map through a reading prediction model, where the reading prediction model is a machine learning model.

[0033] For more information on the functions performed by the processing component 130, see Figure 2-Figure 4 Related description.

[0034] In some embodiments, an inductive sensor testing system may include a movable control console, a sensor panel, a test socket, and a probe. The test socket and probe are positioned below the movable control console and are communicatively connected to a processing assembly. The sensor panel is positioned above the movable control console and may be provided with at least one set of sensors to be tested, with the front faces of the sensors facing upward and their back faces contacting the test probes.

[0035] In some embodiments, an adjustable light source is provided above the sensor panel. When the adjustable light source is turned on, the processing component controls the movable control console to descend, causing the test probe to contact the bottom pads of at least one group of sensors to be tested on the sensor panel. The sensors to be tested that are facing the adjustable light source are tested, and the processing component collects test data. After completing these steps, the processing component controls the movable control console to ascend and move, causing the adjustable light source to face an untested sensor. The movable control console is then controlled to descend again, repeating the above testing process until all sensors to be tested on the sensor panel have been tested.

[0036] In some embodiments, a gray card may be placed at a predetermined distance above the sensor panel. The processing component controls the movable control console to descend, causing the test probe to contact the bottom pads of at least one group of sensors to be tested on the sensor panel. The sensors to be tested that are facing the gray card are tested, and the processing component collects test data. After completing the above steps, the processing component controls the movable control console to ascend, controls the movable control console to move, causes the gray card to face the untested sensors, controls the movable control console to descend again, and repeats the above testing process until all sensors to be tested on the sensor panel are tested.

[0037] It should be noted that the above description of the inductive sensor test system and its modules is for convenience only and does not limit this specification to the scope of the embodiments. It is understandable that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the modules or form subsystems connected with other modules without departing from the principles. In some embodiments, Figure 1 The light shielding box 110, communication component 120, and processing component 130 disclosed herein may be separate modules within a single system, or a single module may implement the functions of two or more of the aforementioned modules. For example, the modules may share a single storage module, or each module may have its own storage module. Such variations are within the scope of this specification.

[0038] Figure 2 FIG. 1 is an exemplary flow chart of an inductive sensor testing method according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps. In some embodiments, the following steps may be performed by the processing component 130 .

[0039] Step 210 : In response to the adjustable light source being activated according to the target parameters, a first reading of at least one standard sensor and a second reading of at least one sensor to be tested are obtained.

[0040] The target parameter is a lighting parameter of the adjustable light source related to the test requirements. In some embodiments, the target parameter may include at least one of the illumination type and the illumination intensity.

[0041] The light intensity refers to the light intensity of the test light emitted by the adjustable light source 111 irradiating the standard sensor and the sensor to be tested.

[0042] The illumination type refers to the type of test light emitted by the adjustable light source 111. In some embodiments, the illumination type may include the wavelength of the light.

[0043] In some embodiments, the target parameters can be determined based on user needs and obtained through manual pre-setting. For example, in order to verify whether the sensor can accurately detect the light intensity of a light source with a wavelength of λ and an intensity of θ, the user needs to set the light source target parameters to wavelength λ and intensity θ.

[0044] The sensor under test refers to the sensor that needs to be tested for reading accuracy.

[0045] In some embodiments, the processing component can control the adjustable light source to emit test light and obtain at least one first reading reflecting the intensity of the test light generated by at least one standard sensor in the light shielding box, as well as at least one second reading reflecting the intensity of the test light generated by at least one sensor under test. The at least one first reading and at least one second reading are used to complete the test of the sensor under test and determine the accuracy of its reading. For more information, please refer to this specification. Figure 4 See the relevant instructions in .

[0046] The first reading refers to a reading of a standard sensor. In some embodiments, the first reading can be a light intensity value.

[0047] The second reading refers to the reading of the sensor to be measured. In some embodiments, the second reading can be a light intensity value.

[0048] In some embodiments, the processing component 130 can communicate with the standard sensor and the sensor under test respectively through the communication component 120 to obtain the first reading and the second reading. In some embodiments, the standard sensor and the sensor under test can send the readings based on actual testing to the processing component through the communication component.

[0049] Step 220 : Determine the reading accuracy of at least one sensor under test based on the first reading and the second reading.

[0050] Reading accuracy refers to the accuracy of the readings of the sensor under test. In some embodiments, reading accuracy can determine whether the actual readings of the sensor under test meet the standard. Sensors under test with low reading accuracy will be notified to the user by the processing component.

[0051] In some embodiments, the processing component can determine the reading accuracy of each sensor under test based on the at least one first reading, the target parameter, and the at least one second reading in a variety of ways.

[0052] For example, for any sensor under test, the processing component can select a standard sensor closest to the sensor under test and obtain a first reading I1 of the standard sensor, a second reading I2 of the sensor under test, a distance R1 between the standard sensor and the light source, and a distance R2 between the sensor under test and the light source. The processing component can calculate the reading accuracy based on the first reading I1 of the standard sensor, the second reading I2 of the sensor under test, the distance R1 between the standard sensor and the light source, and the distance R2 between the sensor under test and the light source.

[0053] As an example only, the reading accuracy can be obtained by calculation based on the second reading of the sensor to be tested and the theoretical reading of the sensor to be tested. For example, the reading accuracy of the sensor to be tested satisfies the following formula (1):

[0054] Reading accuracy =

[0055] Where I2 is the theoretical reading of the sensor under test, and I2' is the second reading of the sensor under test. Formula (1) calculates the reading accuracy by the difference between the theoretical reading and the second reading.

[0056] As an example only, the theoretical reading I2 can be obtained by calculation based on the first reading of the standard sensor, the distance between the standard sensor and the light source, and the distance between the sensor to be tested and the light source. For example, the theoretical reading of the sensor to be tested satisfies formula (2):

[0057]

[0058] Where I1 is the first reading of the standard sensor, R1 is the distance between the standard sensor and the light source, and R2 is the distance between the sensor to be tested and the light source. Formula (2) is obtained by the inverse law of light intensity attenuation, which can be expressed as .

[0059] The above formulas (1) and (2) are merely examples and do not limit the method for calculating the reading accuracy and the theoretical reading of the sensor under test. Other methods for determining the reading accuracy based on the positions of the standard sensor and the sensor under test, as well as the readings of the standard sensor and the sensor under test, can also be applied to this embodiment.

[0060] In some embodiments, the processing component can also determine a theoretical reading in a preset manner based on the first reading, setting data of at least one standard sensor, position distribution data of at least one sensor to be tested, and target parameters, and determine the reading accuracy based on the theoretical reading and the second reading.

[0061] In some embodiments, the processing component may also determine a theoretical reading through a reading prediction model based on the first reading, setting data of at least one group of standard sensors, position distribution data of at least one group of sensors to be tested, and target parameters.

[0062] More information on determining theoretical readings through pre-set methods and reading prediction models can be found in Figure 4 and related instructions.

[0063] Step 230: In response to at least one reading accuracy meeting a warning condition, a prompt message is issued.

[0064] A warning condition refers to a condition used to determine whether to alert a user of low sensor quality. In some embodiments, the warning condition may include a reading accuracy less than an accuracy threshold. The accuracy threshold may be manually preset.

[0065] The prompt information refers to information used to remind the user that the sensor quality is too low. In some embodiments, the prompt information may include, for example, one or more combinations of numbers, text, information, voice, etc.

[0066] In some embodiments, a prompt message may be sent to the user via the communication component to prompt the user that the quality of the sensor to be tested is low.

[0067] In some embodiments, in response to the reading accuracy of the sensor to be tested being less than the accuracy threshold and within the accuracy tolerance, the processor may determine the sensor to be tested as a sensor to be calibrated and calibrate the sensor to be calibrated based on preset calibration rules.

[0068] The accuracy tolerance refers to the permissible range within which the reading of the sensor under test deviates from the first reading. In some embodiments, the accuracy tolerance can be determined based on prior experience and / or actual needs. By way of example only, a standard tolerance may be 20%. That is, if the reading of the sensor under test deviates from the first reading by within 20%, the processor may determine the sensor under test as a sensor to be calibrated and calibrate the sensor under calibration based on preset calibration rules, so that its reading accuracy is no less than an accuracy threshold.

[0069] The preset calibration rule refers to a preset strategy for calibrating the sensor to be calibrated. In some embodiments, the preset calibration rule can be determined based on prior experience and / or actual needs. As an example only, the preset calibration rule may include: adjusting the parameters of the sensor to be calibrated, such as gain, offset, etc., based on the comparison result of the first reading and the second reading, so that the reading of the sensor to be calibrated is as close to the first reading as possible. In some embodiments, the processor can adjust the parameters of the sensor to be calibrated according to the preset adjustment range, and obtain the reading of the sensor to be calibrated after the parameter adjustment. If the accuracy of the reading is less than the accuracy threshold, the adjustment is continued. If the accuracy of the reading is not less than the accuracy threshold, the calibration is stopped, and the adjusted sensor to be calibrated is determined to be a qualified sensor.

[0070] In some embodiments, the difference between the number of standard sensors and the number of sensors to be tested is greater than a preset value, wherein the preset value can be determined based on the illumination intensity of the adjustable light source.

[0071] As an example only, the preset value can be calculated based on the light intensity of the target parameter and the recognition range of the sensor. The sensor recognition range refers to the range of the measured value within which the sensor's error is within a specified limit. In some embodiments, the error within the sensor recognition range can be ignored. For example, the preset value satisfies the following formula (3):

[0072] Default value = *Standard presets (3)

[0073] Where x is the light intensity of the target parameter, a is the lower limit of the sensor recognition range, and b is the upper limit of the sensor recognition range.

[0074] In some embodiments, the greater the sensor detection error, the more standard sensors are required.

[0075] As you can understand, the further the target parameter's light intensity is from the edge of the sensor's recognition range, the more accurate the sensor's detection will be, and the number of standard sensors can be reduced. The closer the target parameter's light intensity is to the edge of the sensor's recognition range, the greater the sensor's detection error will be, and more standard sensors will be needed to achieve more accurate detection results.

[0076] In some embodiments of the present specification, the inductive sensor testing system arranges multiple standard sensors and multiple sensors to be tested in a light-shielding box, and enables an adjustable light source to provide illumination under the same target parameters. The system can obtain the first reading and the second reading corresponding to each of the standard sensors and the sensors to be tested through a single illumination, judge the quality of the sensors to be tested by calculating the accuracy of the readings, and test multiple sensors to be tested at one time, thereby reducing possible errors in sensor testing and improving test accuracy and efficiency.

[0077] It should be noted that the above description of process 200 is for illustration and purpose only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and variations to process 200 under the guidance of this specification. However, such modifications and variations are still within the scope of this specification.

[0078] Figure 3 is an exemplary schematic diagram of determining setting data of a standard sensor according to some embodiments of this specification.

[0079] In some embodiments, the processing component may determine the setting data 370 of the at least one set of standard sensors by referring to a database 360 according to the target parameter 310 and the position distribution data 320 of the at least one set of sensors to be tested in the light shielding box.

[0080] The position distribution data reflects the position distribution of the sensors to be tested in the light shielding box.

[0081] In some embodiments, the position distribution data can be represented by a position distribution coordinate set. The position distribution coordinate set includes the coordinates of several sensors under test. For example, the processing component can construct a three-dimensional spatial coordinate system based on a preset origin, using the coordinates of the sensors under test within the light shielding box as elements of the position distribution coordinate set. The preset origin can be predetermined based on prior experience. For example, the geometric center of the light shielding box can be determined as the preset origin.

[0082] In some embodiments, location distribution data may be determined based on target parameters and user testing requirements.

[0083] User test requirements are test requirements set by users for the sensor to be tested. In some embodiments, user test requirements may include the received light intensity of the sensor to be tested, the distance between the sensor to be tested and the adjustable light source, etc. For example, when the user test requirement is to detect a target parameter of 100Lux, whether the sensor to be tested 5m away can accurately detect the light intensity, the processing component can determine several positions 5m away from the adjustable light source as target positions, and determine the position distribution data based on the aforementioned target positions. For another example, when the user test requirement is to detect whether the reading of the sensor to be tested is accurate when the received light intensity is 50Lux, the processing component can calculate the position that can make the received light intensity of the sensor to be tested 50Lux based on methods such as the light attenuation equation, and determine the coordinates corresponding to the position as the position distribution data. For more explanation of the light attenuation equation, please refer to Figure 2 It is understandable that when setting the position of the sensors to be tested, each sensor to be tested should be placed at a position where the test light emitted by the adjustable light source can directly illuminate it.

[0084] The setting data 370 may reflect the position distribution of the standard sensors within the light shielding box.

[0085] In some embodiments, the position distribution data may be represented by a set of coordinates. The configuration of the set of coordinates is similar to that of the position distribution coordinates, and reference may be made to the above description.

[0086] In some embodiments, the setting data may be determined by reference to a database.

[0087] The reference database is a database used to determine the setting data. In some embodiments, the reference database includes at least one reference vector. The reference vector may include a first reference vector 310 and a second reference vector 320.

[0088] The first reference vector is one of reference vectors used to determine the spatial distribution of the standard sensor.

[0089] The first reference vector includes at least one position element representing the position of the standard sensor and a parameter element representing the reference target parameter. In some embodiments, the first reference vector corresponds to a situation where the standard sensors are irregularly distributed within the light shielding box.

[0090] In some embodiments, the processing component may determine at least one first reference vector based on gold standard simulation data.

[0091] Gold standard simulation data refers to test-related data obtained by performing simulation tests on standard sensors.

[0092] In some embodiments, the gold standard simulation data may include reference position distribution data of a standard sensor, test readings, and reference target parameters used for simulation when performing a simulation test.

[0093] In some embodiments, the processing component can randomly generate multiple sets of reference position distribution data, each set of reference position distribution data corresponding to at least one reference target parameter. Multiple sets of to-be-simulated data are generated by randomly combining the multiple sets of reference position distribution data and the at least one reference target parameter, each set of to-be-simulated data including a set of position distribution data and a reference target parameter.

[0094] In some embodiments, the number of standard sensors in each set of reference position distribution data can be determined based on a preset number range. For example, the preset number range can be 20-50. In this case, the number of standard sensors in the generated multiple sets of reference position distribution data can be any value between 20-50. The number of groups with different numbers of standard sensors is evenly distributed.

[0095] In some embodiments, each piece of gold standard simulation data has a corresponding label, which may be the number of the standard sensor used for the simulation test.

[0096] In some embodiments, the reference target parameters may be determined based on historical experience or actual demand. For example, the processing component may determine the reference target parameters based on at least one historical actual target parameter. For another example, the processing component may determine new reference target parameters based on actual demand.

[0097] In some embodiments, the processing component can generate as much reference position distribution data and reference target parameters as possible to ensure that the number of samples for the simulation test is large enough, thereby facilitating the screening and acquisition of sufficient target data.

[0098] In some embodiments, the processing component can perform simulation tests on the aforementioned multiple sets of simulated data using various simulation methods to obtain test readings of the standard sensor under different reference target parameters. Gold standard simulation data is determined based on the simulated data and its corresponding test readings. Simulation methods may include, but are not limited to, simulation experiments and computer simulations.

[0099] In some embodiments, the processing component may screen the aforementioned gold standard simulation data and determine the gold standard simulation data whose test results meet the retention conditions as target data.

[0100] In some embodiments, the retention condition may include that the relative error of the test is less than a preset retention threshold. The smaller the relative error, the better the effect of the test under the gold standard simulation data.

[0101] In some embodiments, for a gold standard simulation data, the processing component can randomly select several standard sensors and their corresponding test readings from multiple standard sensors, and determine the test evaluation value corresponding to the standard sensor based on the distance between the aforementioned several standard sensors and the test light source and the test readings.

[0102] For example, the processing component may record the ratio of the distance of each standard sensor from the test light source to the square of the test reading as the test evaluation value S. Statistical variance of the test evaluation value S of at least one standard sensor , the variance It is recorded as relative error. If the value is less than the preset retention threshold, the gold standard simulation data is considered to meet the test requirements and the gold standard simulation data can be retained. The preset retention threshold can be preset based on prior experience.

[0103] For example, the processing component can randomly select 20 standard sensors from a gold standard simulation data, and record the distance to the test light source and the test reading as (L1, R1), (L2, R2), ..., (L20, R20), respectively. Among them, the first element in the vector refers to the distance, and the second element is the test reading. The processing component can calculate the test evaluation value of each standard sensor, such as L1 / (R1^2), recorded as S1; calculate L2 / (R2^2), recorded as S2; and so on, obtain S1, S2, ..., S20 and calculate the variance to obtain the relative error corresponding to the gold standard simulation data. . judge Is it less than the preset retention threshold? If so, it is determined to retain the gold standard simulation data.

[0104] For another example, the processing component can randomly select different numbers of standard sensors multiple times, and calculate the relative error based on the above method multiple times based on the distance between the selected standard sensors and the test light source and the test readings. , in response to the relative error If the average value of is less than the preset retention threshold, the gold standard simulation data is retained.

[0105] Through the above method, the processing component can determine at least one target data. In some embodiments, the processing component can construct a first vector based on the target data. For example, based on a preset encoding rule and an encoding upper limit, the position coordinates of the standard sensor in the target data are encoded; the encoded position coordinates are sorted according to the encoding value, and combined with the reference target parameters contained in the target data to construct a first reference vector.

[0106] In some embodiments, the first reference vector has a label, and the label can be determined based on the label of the corresponding target data.

[0107] In some embodiments, the processing component may encode the placement positions of the sensors in the light shielding box using encoding rules, so that each position corresponds to a unique encoding value.

[0108] In some embodiments, the encoding value can be a positive integer. For example, for a position to be encoded, the processing component can obtain the coordinates of the position in a three-dimensional space coordinate system and encode it based on an encoding rule. As an example, the encoding rule can be: sort the coordinates from small to large according to the value of the X-axis; for coordinates with the same X-axis, sort the Y-axis from small to large; for coordinates with the same X-axis and Y-axis, sort the Z-axis from small to large.

[0109] After the sorting is complete, the first-order coordinate (i.e., the one with the smallest X-axis value) is encoded as 1, the second-order coordinate is encoded as 2, and so on. When a coordinate that needs to be encoded appears again, the processing component can compare the coordinate with the previously encoded coordinate. If a code exists for the coordinate, the existing code is used. If no corresponding code exists for the coordinate, the processing component can sort the coordinates from small to large along the X-axis, Y-axis, and Z-axis, select the smallest code value that does not yet have a corresponding coordinate as the encoding starting value, and encode according to the sorting.

[0110] For example, target data 1 includes three standard sensors, with corresponding coordinates: [(0.5, 0.5, 0.5), (0.5, 1, 0.5), (1, 0.5, 1)]. The processing component can sort the coordinates in ascending order from the X-axis, Y-axis, and Z-axis, encoding coordinate (0.5, 0.5, 0.5) as 1, coordinate B (0.5, 1, 0.5) as 2, and coordinate (1, 0.5, 1) as 3. Target data 2 includes five standard sensors, with corresponding coordinates: [(0.5, 0.5, 0.5), (0.5, 1, 0.5), (1, 0.5, 1), (0.5, 1, 1), (1, 2.5, 2)]. Coordinates (0.5, 0.5, 0.5), (0.5, 1, 0.5), and (1, 0.5, 1) have already been encoded as 1, 2, and 3, so the previous encoding is used directly. The coordinates (0.5, 1, 1) and (1, 2.5, 2) have not appeared before and are not encoded. Sort them from small to large along the X-axis, Y-axis, and Z-axis, and encode the coordinate (0.5, 1, 1) as 4 and the coordinate (1, 2.5, 2) as 5. This continues in this way, encoding all positions where standard sensors can be placed.

[0111] In some embodiments, the processing component can construct an initial first reference vector based on the codes corresponding to the standard sensor positions in the target data and the target parameters. For example, if the codes corresponding to the standard sensor positions in target data 1 are 1, 2, and 3, and the target parameter is luminous intensity 400 Lux, the corresponding first reference vector is (1, 2, 3; 400 Lux).

[0112] In some embodiments, to facilitate comparisons and calculations between different vectors, the processing component may pad the initial first reference vector based on the coding upper limit so that all first reference vectors contain the same number of elements. The padding method may be: first, add several zero elements to the end of the initial first reference vector to pad the number of elements in the first reference vector to the coding upper limit W + ΔW. Then, reorder the elements in the initial first reference vector, move the element coded as n to the nth position of the vector, and fill the remaining positions with zero elements to obtain the first reference vector.

[0113] For example, if the current encoding limit is 10, the initial first reference vector corresponding to standard data 2 is (1, 3, 4, 7, 9; 400 Lux). First, add several zero-position elements to the vector to padded the vector, resulting in (1, 3, 4, 7, 9, 0, 0, 0, 0; 400 Lux). Then, reorder the vector, moving the elements encoded as 1, 3, 4, 7, and 9 to the 1st, 3rd, 4th, 7th, and 9th positions of the vector, respectively. The remaining positions are padded with zero elements, resulting in the first reference vector (1, 0, 3, 4, 0, 0, 7, 0, 9, 0; 400 Lux).

[0114] In some embodiments, the processing component may count the total number W of different emission positions of the standard sensor in the gold standard simulation data whose test results meet the requirements, and set W + ΔW as the encoding upper limit of the position element of the first reference vector. ΔW is the element threshold, and the value of ΔW can be determined based on prior experience.

[0115] The second reference vector is one of the reference vectors used to determine the spatial distribution of the standard sensor.

[0116] In some embodiments, the second reference vector includes at least one position element representing the position of the standard sensor and a parameter element representing a reference target parameter. In some embodiments, the second reference vector corresponds to a situation where the standard sensors are neatly arranged in the light shielding box.

[0117] In some embodiments, the processing component can also grid the shading box based on the shading parameters to obtain at least one grid; evaluate the credibility 330 of at least one grid based on the target parameters and position distribution data; and determine the at least one second reference vector 320 based on the light loss 350 of at least one grid, the number 340 of the standard sensors and the credibility 330.

[0118] In some embodiments, the second reference vector may have a label, and the label may be determined based on the number of standard sensors in the second reference vector.

[0119] The shading parameters refer to parameters that can reflect the shading properties of the shading box. In some embodiments, the shading parameters may include the length, width, and height of the interior of the shading box. In some embodiments, the shading parameters can be obtained by reading the specification data of the shading box and / or by obtaining manual input.

[0120] In some embodiments, the processing component may perform gridding on the light shielding box based on the light shielding box's light shielding parameters and grid parameters to obtain at least one grid. The grid parameters may include a grid shape (e.g., a cuboid, a cube, etc.) and grid dimensions (e.g., length, width, and height).

[0121] In some embodiments, the maximum number of sensors within each grid is one. The grid parameters can be preset based on prior experience, and the grid size should be at least larger than the average size of the sensors. In some embodiments, the processing component can model the interior space of the light shielding box in a three-dimensional coordinate system according to the light shielding parameters, and divide the model based on the grid parameters to obtain a plurality of grids having grid shapes and grid sizes.

[0122] In some embodiments, the processing component can screen the grids and exclude the grids located in the shadow of the sensor, thereby ensuring that the light intensity in the remaining grids is not affected by the sensor. The processing component can randomly select a preset number of grids whose grid credibility meets the requirements from the remaining grids as distributable grids that can place the standard sensor, select a position for placing the standard sensor from each distributable grid, and obtain the coordinates of the position, encode the position coordinates with reference to the encoding rules, and use the encoded value as the position element of the initial second reference vector. The corresponding target parameter is used as the target parameter element of the initial second reference vector. The initial second reference vector is padded to obtain the second reference vector. By continuously repeating the above steps, several second reference vectors can be obtained.

[0123] The grid credibility requirement can be satisfied by the grid credibility being greater than a preset grid credibility threshold. The preset distributable number is positively correlated to the number of standard sensors. The specific steps for constructing and completing the second reference vector are similar to those for the first reference vector, and can be found in the relevant content above.

[0124] The credibility can reflect the reliability of the readings of the sensors to be tested within a certain grid. In some embodiments, the credibility of a grid can be represented by a numerical value, where the closer the value is to 1, the higher the reliability of the current grid.

[0125] In some embodiments, the processing component can determine the credibility in a variety of ways.

[0126] In some embodiments, the processing component may determine the credibility of a grid based on a first preset table. The processing component may construct the first preset table based on historical target parameters, historical location distribution data, historical grid locations, and historical credibility. The first preset table contains correspondences between different historical target parameters, historical location distribution data, historical grid locations, and different historical credibility levels. The processing component may determine the current credibility based on the current target parameters, location distribution data, and grid location by querying the first preset table. The grid location may be the coordinates of the geometric center point of the grid.

[0127] In some embodiments, the reliability of a grid can be determined based on the light intensity at key locations associated with the grid, wherein the key locations include the center of the grid and the location of the sensor to be tested closest to the grid.

[0128] In some embodiments, the processing component may select a sensor to be tested that is closest to the geometric center of the grid, calculate a first light intensity at the center point of the sensor to be tested, and calculate a second light intensity at the center point of the grid; and determine credibility based on a ratio of the second light intensity to the first light intensity.

[0129] In some embodiments, the illumination intensity of the sensor to be tested or the center point of the grid is positively correlated with the illumination intensity of the adjustable light source, and negatively correlated with the distance from the center point of the sensor to be tested or the center point of the grid to the adjustable light source.

[0130] In some embodiments, the processing component can calculate the light intensity L at the center point of the sensor to be measured or the center point of the grid by formula (4):

[0131]

[0132] Where L represents the light intensity at the center of the sensor under test or the center of the grid, G represents the light intensity of the adjustable light source, R represents the distance from the center of the sensor under test or the center of the grid to the adjustable light source, and μ is a coefficient. The coefficient μ can be preset based on prior experience. The processing component can obtain the coordinates of the center of the sensor under test or the center of the grid and calculate the distance R between the two coordinates using the Euclidean distance formula.

[0133] In some embodiments of the present specification, the internal space of the light shielding box is divided into grids so that there can be at most one sensor in each grid, thereby ensuring that the sensors are not arranged too closely and thus affect each other, and ensuring that the sensors can be arranged in an orderly and neat manner for easy installation and calculation; the second standard vector can be supplemented according to the current situation, which on the one hand reduces the amount of calculation, and on the other hand can make the vectors in the standard vector library more targeted, which can improve the calculation speed during matching.

[0134] In some embodiments, the grid credibility is also related to the adjustable light source position and at least one relative position. The processing component can evaluate the light loss degree of at least one grid through the light loss degree prediction model based on the target parameters, position distribution data, the adjustable light source position and the relative position, and evaluate the grid credibility of at least one grid based on the light loss degree of at least one grid. For an explanation of the target parameters, please refer to this specification. Figure 2 For the relevant description in , please refer to the previous description for the description of position distribution data and adjustable light source position.

[0135] The relative position refers to the position of the grid relative to the light box. In some embodiments, the relative position can be represented by a vector. The starting point of the vector can be the geometric center of the light box, and the end point of the vector can be the geometric center of the grid.

[0136] Light loss refers to the loss of light emitted by an adjustable light source during its propagation. Examples include diffuse reflection and absorption by dust and impurities in the air. This loss can cause the actual light intensity received by the sensor to be less than the theoretical light intensity. In some embodiments, light loss can be expressed as a percentage of the light loss during propagation.

[0137] The optical loss degree of the grid refers to the loss degree of the test light emitted by the adjustable light source when it reaches the position of the grid.

[0138] The light loss prediction model is used to determine light loss. In some embodiments, the light loss prediction model can be a machine learning model, such as a neural network (NN) model or any combination thereof, or other custom model structures.

[0139] In some embodiments, the input of the light loss degree prediction model may include target parameters, position distribution data, adjustable light source positions, and relative positions of at least one grid, and the output may include the light loss degree of at least one grid.

[0140] In some embodiments, the processing component may obtain a light loss degree prediction model through training based on a large number of first training samples with first labels. In some embodiments, the first training samples may include sample target parameters, sample position distribution data, sample adjustable light source positions, and sample relative positions. The first label may be the sample light loss degree of at least one grid.

[0141] In some embodiments, the first training sample can be obtained based on historical data. The actual optical loss level corresponding to at least one grid can be obtained by using an instrument such as an optical loss detector, and used as the first label corresponding to the first training sample.

[0142] In some embodiments, the processing component can be trained using various methods based on the first training samples and the first label. For example, training can be performed based on a gradient descent method. As an example, multiple first training samples with first labels can be input into a light loss prediction model. A loss function can be constructed using the first labels and the results of the light loss prediction model. The parameters of the light loss prediction model can then be iteratively updated based on the loss function. Model training is completed when the loss function of the light loss prediction model meets preset conditions, resulting in a trained missed detection degree determination model. The preset conditions can include, for example, convergence of the loss function or a threshold number of iterations.

[0143] In some embodiments, the grid credibility may be determined based on the light loss corresponding to the grid, the light intensity at the center point of the grid, and the distance between the center point of the grid and the adjustable light source.

[0144] In some embodiments, the grid credibility D can be calculated by formula (5):

[0145]

[0146] Where D represents the grid reliability, F represents the light loss corresponding to the grid, L2 represents the light intensity at the center of the grid, R represents the distance between the center of the grid and the adjustable light source, and μ is a coefficient. The coefficient μ can be preset based on prior experience.

[0147] Some embodiments of this specification determine light loss using a light loss degree prediction model, which can fully utilize the model's data processing and analysis capabilities, comprehensively consider the impact of multiple factors on the light loss degree, and thus improve the accuracy and reliability of light intensity prediction.

[0148] The vector to be matched is a vector used to represent the position distribution of the sensor to be tested. In some embodiments, the vector to be matched may include several position elements and a target parameter element, each position element corresponding to the position of a sensor to be tested in the light shielding box, and the target parameter element corresponding to the target parameter.

[0149] In some embodiments, the processing component can determine the coordinates of the sensor to be tested based on the position distribution data of the sensor to be tested, encode the coordinates according to the aforementioned encoding rules, use the encoded results as the position elements of the initial vector to be matched, and use the corresponding target parameters as the target parameter elements to pad the initial vector to be matched to obtain the vector to be matched. The padding method can be found in the relevant description above.

[0150] In some embodiments, the processing component may perform similarity matching between the vector to be matched and candidate reference vectors, select the candidate reference vector with the highest similarity as the target reference vector, and determine the setting data of at least one standard sensor based on the target reference vector and the vector to be matched. Similarity matching algorithms may include, but are not limited to, Manhattan distance methods and support vector machines.

[0151] The candidate reference vector may be a reference vector whose tag satisfies a selection condition, wherein the selection condition may be that the tag value is equal to the number of sensors in the vector to be matched, wherein the number of sensors in the vector to be matched is the sum of the number of standard sensors and the number of sensors to be tested.

[0152] In some embodiments, the processing component can calculate the absolute value of the difference between the target reference vector and the vector to be matched, and determine the elements in the absolute value that are not zero and are not light source parameters as the setting data. For example, the vector to be matched is (1, 0, 0, 4, target parameter 5), and the target reference vector is (1, 2, 3, 4, target parameter 6); the absolute value of the difference is (0, 2, 3, 0, |target parameter 5 - target parameter 6|), and the elements in the absolute value that are not zero and are not light source parameters are determined to be the position distribution (2, 3). During testing, the standard sensor can be placed at point B, which is coded as 2, and point C, which is coded as 3.

[0153] In some embodiments of the present specification, a reference database is constructed based on gold standard simulation data, and the required first vector is determined through similarity matching, thereby determining the placement position of the standard sensor. This can effectively ensure that the relative positions of the standard sensor and the sensor to be tested are basically consistent, avoid deviations due to unreasonable position distribution, and improve the accuracy and reliability of sensor monitoring.

[0154] Figure 4 is an exemplary schematic diagram of determining reading accuracy according to some embodiments of this specification.

[0155] In some embodiments, the processing component can determine the theoretical reading 450 of the sensor to be tested based on the first reading 410, the setting data 420 of at least one standard sensor, the position distribution data 320 of at least one sensor to be tested, and the target parameter 310; and determine the reading accuracy 470 based on the theoretical reading and the second reading 460.

[0156] The reading accuracy can reflect the accuracy of the light intensity measurement results of the sensor under test. In some embodiments, the reading accuracy can be negatively correlated with the reading difference value. The reading difference value refers to the difference between the second reading corresponding to the sensor under test and its theoretical reading.

[0157] In some embodiments, the reading accuracy corresponding to the sensor to be tested can be calculated using formula (6):

[0158]

[0159] Where η represents the reading accuracy, G represents the second reading corresponding to the sensor under test, and V represents the theoretical reading corresponding to the sensor under test. The second reading corresponding to the sensor under test can be obtained based on the test results. For information on obtaining the theoretical reading, please refer to the following.

[0160] The theoretical reading of the sensor to be tested refers to the reading of the sensor to be tested under an ideal state. In some embodiments, the sensor can determine the theoretical reading in a variety of ways.

[0161] In some embodiments, the processing component can determine a theoretical reading using a preset method based on the first reading and position data of the standard sensor, the position distribution data of the sensor to be tested, and the target parameter. For example, at least one candidate sensor group corresponding to the sensor to be tested is determined from the standard sensors; based on the isotropic reliability of each candidate sensor group, a target sensor group is determined for calculating the theoretical reading of the sensor to be tested; and the theoretical reading of the sensor to be tested is determined based on the reference theoretical reading of the reference sensor in the target sensor group.

[0162] In some embodiments, the processing component may determine candidate reference sensors according to a selection rule based on the configuration data of the standard sensors and the position distribution data of the sensor to be tested. In some embodiments, the selection rule may be: sorting the standard sensors by their distance from the sensor to be tested, from closest to farthest, and selecting the first N standard sensors with the shortest distances as candidate reference sensors.

[0163] In some embodiments, the processing component may randomly select several candidate reference sensors from the candidate reference sensors to form a candidate sensor group; and by repeating the aforementioned random selection steps, several candidate sensor groups may be obtained.

[0164] In some embodiments, the processing component may calculate the isotropic credibility of each candidate sensor group and filter out the target sensor based on the isotropic credibility.

[0165] The isotropy credibility can reflect the overall reliability of the sensor readings in the candidate sensor group. The larger the value of the isotropy credibility, the more reliable the sensor readings included in the candidate sensor group are and the higher the spatial isotropy is.

[0166] In some embodiments, the processing component can select any two candidate reference sensors from the candidate sensor group and obtain their corresponding first readings, and calculate the theoretical relative distance ratio between the two candidate reference sensors and the adjustable light source based on the first readings. , calculate the actual relative distance ratio between the two sensors and the adjustable light source based on the setting data and position distribution data .

[0167] calculate The value of is recorded as ∝.

[0168] Traverse all combinations of two candidate reference sensors from the candidate sensor group to obtain several groups of ∝ values, and calculate the average value of the aforementioned several groups of ∝ values Then the isotropic credibility of the candidate reference sensor group is equal to 1- .

[0169] In some embodiments, the processing component may sort the candidate sensor groups corresponding to the sensor to be tested from high to low according to isotropic credibility, select the candidate sensor group with the highest isotropic credibility, and determine all candidate reference sensors contained therein as target sensors.

[0170] In some embodiments, the processing component can calculate at least one candidate theoretical reading of the sensor to be tested based on the position distribution data of the sensor to be tested, the setting data of at least one target sensor, and the first reading of the standard sensor, and determine the theoretical reading of the sensor to be tested based on the statistical value (e.g., average, median, etc.) of the at least one candidate theoretical reading. For example, the average value of the at least one candidate theoretical reading can be determined as the theoretical reading of the sensor to be tested. The candidate theoretical reading of the sensor to be tested can be obtained based on the inverse square law of light intensity attenuation. For more information about the inverse square law, please refer to Figure 2 See the relevant instructions in .

[0171] Some embodiments of this specification can quantify the detection accuracy of the sensors to be tested by determining the theoretical reading of each sensor to be tested and the corresponding reading accuracy, which helps to evaluate the test effect of each sensor to be tested, thereby improving the consistency and reliability of the inductive sensor testing system.

[0172] In some embodiments, the position distribution data can be determined based on the parameters of the light source to be measured and the user's test requirements, and the setting data can be determined based on the vector matching method of the first reference vector. For relevant descriptions of the parameters of the light source to be measured, the user's test requirements, and the vector matching method, please refer to Figure 2 Related content in .

[0173] In some embodiments, the processing component can construct a prediction map 430 based on the first reading 410, the setting data 420 of at least one standard sensor, the position distribution data 320 of at least one sensor to be tested, and the target parameter 310; based on the prediction map 430, the theoretical reading 450 is determined through the reading prediction model 440.

[0174] The predicted map is a map that shows the distribution of devices within the light shielding box. The predicted map can reflect the distribution characteristics of the standard sensors, sensors under test, and adjustable light sources within the light shielding box.

[0175] In some embodiments, the prediction graph may be composed of at least one node and at least one edge, where the node has corresponding node attributes and the edge has corresponding edge attributes.

[0176] In some embodiments, the node attributes include at least the node coordinates.

[0177] The nodes include a standard node 432, a node to be tested 433, and a light source node 431. The standard node represents a standard sensor, the node to be tested represents a sensor to be tested, and the light source node represents an adjustable light source. The node attributes of the standard node also include a first reading, the node attributes of the sensor node to be tested also include a second reading, and the node attributes of the adjustable light source node also include target parameters.

[0178] In some embodiments, the coordinates of the standard node may correspond to the coordinates of the standard sensor in a three-dimensional coordinate system, the coordinates of the node to be measured may correspond to the coordinates of the standard sensor to be measured in the three-dimensional coordinate system, and the coordinates of the light source node may correspond to the coordinates of the adjustable light source in the three-dimensional coordinate system. The processing component may construct a three-dimensional coordinate system within the light shielding box based on a preset origin, and determine the coordinates of the standard sensor to be measured and the adjustable light source based on the positions of the sensor to be measured and the adjustable light source in the three-dimensional coordinate system.

[0179] In some embodiments, any two nodes may be connected via an edge, and the attributes of the edge may include the distance between the nodes connected by the edge.

[0180] In some embodiments, the edges include first-type edges 435 and second-type edges 434. The first-type edges are used to connect sensors to each other, and the second-type edges are used to connect adjustable light sources to sensors.

[0181] In some embodiments, the edge feature of the first type of edge further includes an influence coefficient.

[0182] The influence coefficient reflects the degree of interference a sensor experiences from other sensors. In some embodiments, light striking a sensor surface may not be completely absorbed, but may be reflected or refracted. Furthermore, sensors may block each other, causing interference and reducing sensor detection accuracy. Therefore, the influence coefficient can be used to measure the degree of this interference.

[0183] The influence coefficient determination model is a prediction model for determining the influence coefficient. In some embodiments, the influence coefficient determination model may be a neural network (NN) model.

[0184] In some embodiments, the input of the influence coefficient determination model may include node attributes corresponding to two nodes connected by a first-type edge, and the output may be the influence coefficient of each sensor.

[0185] In some embodiments, the processing component can train an influence coefficient determination model based on a large number of third training samples with a third label. The processing component can determine the node coordinates and node types corresponding to the two nodes connected by the first type of edge in the prediction map as the third training sample. The third label can be obtained based on a simulation experiment. For example, the test environment of the current prediction map can be simulated in the experiment, and only the sensor 1 corresponding to one node connected by the first type of edge is placed, and the reading A1 of the sensor is obtained. The sensor 2 corresponding to the other connected node is placed, and the reading A2 of the sensor is obtained. Calculate (A1-A2) / A1, and determine the calculation result as the influence coefficient. Repeat the experiment several times, and calculate the average value of the influence coefficients corresponding to all the first type edges of the prediction map to determine it as the third label.

[0186] The specific steps for determining the training influence coefficient are similar to the steps for training the optical loss degree prediction model, and can be found in the relevant content above.

[0187] In some embodiments of this specification, by using an influence coefficient determination model to determine the influence coefficient and including the influence coefficient in the third training sample, the influence coefficient determination model can comprehensively consider the mutual influence between multiple sensors, thereby improving the generalization ability and adaptability of the model.

[0188] The reading prediction model is used to predict the theoretical reading of the sensor under test. In some embodiments, the reading prediction model can be a machine learning model, such as a Graph Neural Network (GNN) model or any combination thereof, or other custom model structures.

[0189] In some embodiments, the input of the reading prediction model may include a prediction map, and the output of the reading prediction model may include a theoretical reading of the sensor to be tested corresponding to the node.

[0190] In some embodiments, the processing component can train a reading prediction model based on a plurality of second training samples with second labels.

[0191] In some embodiments, the processing component can construct a sample prediction graph based on historical first readings, historical setting data, and historical target parameters in historical test data, where the historical test data refers to data obtained based on simulated experiments or actual historical tests.

[0192] In some embodiments, for each second training sample, the processing component may determine historical standard sensors based on the historical setting data, randomly select a first preset number of historical standard sensors from the historical standard sensors, and construct a first preset number of sample nodes in the sample prediction graph. The positions of the sample nodes in the sample prediction graph correspond to the coordinates of the historical standard sensors in the light shielding box.

[0193] In some embodiments, the first preset number corresponding to each second training sample may be the same or different, the first preset number may be selected from a first preset range, the first preset range may be positively correlated with the number of historical standard sensors, and the value of the first preset range may be determined based on prior experience.

[0194] In some embodiments, the processing component can determine historical coordinates and historical target parameters of the adjustable light source based on the historical setting data, construct a corresponding sample light source node in the sample prediction graph, and determine the historical target parameters as sample target parameters of the sample light source node.

[0195] In some embodiments, the processing component may select a second preset number of sample nodes from the sample prediction graph and determine them as sample test nodes. The unselected sample nodes may be determined as sample standard nodes. The second preset number may be preset based on prior experience.

[0196] In some embodiments, the second preset numbers corresponding to each second training sample may be the same or different.

[0197] In some embodiments, when generating the second training sample, the processing component may perform random sampling from at least one preset proportion and determine the second preset number based on the result of the random sampling.

[0198] The preset proportion refers to the percentage of the preset second preset quantity to the first preset quantity. In some embodiments, the processing component can determine at least one preset proportion by random generation, based on manual input and / or based on historical test data.

[0199] In some embodiments, at least one preset proportion has certain distribution characteristics, such as a degree of discreteness.

[0200] In some embodiments, the distribution characteristic of the at least one preset ratio can be represented by a variance. The larger the variance, the more dispersed the distribution of the at least one preset true ratio.

[0201] In some embodiments, the variance of the at least one preset proportion is related to the size parameter of the light shielding box. For example, the variance may be positively correlated to the size parameter of the light shielding box, and the larger the volume of the light shielding box, the larger the variance of the at least one preset proportion.

[0202] In some embodiments, when determining the second preset quantity, the processing component may perform a random sampling from at least one preset proportion to obtain at least one target proportion, and determine at least one second preset quantity based on the target proportion and the first preset quantity.

[0203] Because the larger the light shielding box, the greater the instability of the test results, some embodiments of this specification can, when constructing the second training sample, increase the variance of at least one preset proportion for larger light shielding boxes to ensure the diversity of the second training sample and the stability of the test results.

[0204] In some embodiments, the processing component can determine node attributes for a sample node to be tested and a sample standard node. For the sample node to be tested, the historical first reading of the corresponding historical standard sensor is determined as the sample second reading; for the sample standard node, the historical first reading of the corresponding historical standard sensor is determined as the sample first reading. The coordinates of the node's corresponding historical standard sensor are determined as the coordinates of the sample node to be tested and the sample standard node.

[0205] In some embodiments, the processing component can determine a sample edge and its corresponding sample edge attributes. The processing component can connect any two sample nodes and, based on the types of the two sample nodes (sample standard node, sample test node, or sample light source node), determine whether the sample edge is a first-class edge or a second-class edge. The edge attributes can also be determined based on the path distance between the two sample nodes.

[0206] In some embodiments, the processing component may generate a required number of sample prediction graphs as second training samples by repeating the above steps.

[0207] In some embodiments, the second label may be a sample second reading of the sample node to be tested corresponding to the second training sample. The sample second reading may be determined based on historical second readings in historical data.

[0208] The specific steps for training the reading prediction model are similar to those for training the light loss prediction model, and can be found in the relevant content above.

[0209] In some embodiments of this specification, a prediction graph is constructed as input to a machine learning model to determine a theoretical reading, which can fully utilize the data processing and data analysis capabilities of the machine learning model to quickly and accurately obtain the desired theoretical reading.

[0210] While the basic concepts have been described above, it will be apparent to those skilled in the art that the detailed disclosure is merely illustrative and does not limit this specification. Although not explicitly stated herein, various modifications, improvements, and revisions to this specification may be made by those skilled in the art. Such modifications, improvements, and revisions are suggested in this specification and remain within the spirit and scope of the exemplary embodiments of this specification.

[0211] This specification also uses specific terms to describe the embodiments of this specification. For example, "one embodiment," "an embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "one embodiment," "an embodiment," or "an alternative embodiment" two or more times in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of this specification may be appropriately combined.

[0212] In addition, unless expressly stated in the claims, the order of the processing elements and sequences, the use of alphanumeric characters, or the use of other names described in this specification are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the spirit and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0213] Similarly, it should be noted that, in order to simplify the presentation of this specification and thus facilitate understanding of one or more embodiments of the invention, the foregoing descriptions of the embodiments of this specification sometimes combine multiple features into a single embodiment, figure, or description thereof. However, this disclosure method does not imply that the subject matter of this specification requires more features than those recited in the claims. In fact, an embodiment may have fewer features than all of the features of a single disclosed embodiment.

[0214] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise stated, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the description and claims are approximate values, which may vary according to the required features of the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the settings of such numerical values are as accurate as possible within the feasible range.

[0215] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, and documents, cited in this specification is hereby incorporated by reference in its entirety. This excludes any application history documents that are inconsistent with or conflicting with the content of this specification, as well as any documents (currently or subsequently appended to this specification) that limit the broadest scope of the claims of this specification. It should be noted that if the descriptions, definitions, and / or terminology used in the accompanying materials are inconsistent or conflicting with the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0216] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. An inductive sensor testing system, characterized in that: Includes light shielding box, communication components and processing components; An adjustable light source, at least one set of standard sensors and a slide rail device are provided in the light shielding box, and at least one set of sensors to be tested are placed in the light shielding box for testing; The adjustable light source emits non-parallel light into the light shielding box; The adjustable light source, the at least one group of standard sensors, and the at least one group of sensors to be tested move along the slide rail device; The processing component is communicatively connected with the adjustable light source, the at least one group of standard sensors, and the at least one group of sensors to be tested respectively through the communication component; The processing component is configured to: In response to the adjustable light source being activated according to a target parameter, obtaining a first reading of the at least one set of standard sensors and a second reading of the at least one set of sensors to be tested; the target parameter includes the light intensity of the adjustable light source; According to the target parameters and the position distribution data of the at least one group of sensors to be tested in the light shielding box, setting data of the at least one group of standard sensors is determined by referring to a database, wherein the setting data represents the position distribution of the at least one group of standard sensors in the light shielding box; The reference database includes at least one first reference vector, which is one of the reference vectors used to determine the spatial distribution of the standard sensor, and the first reference vector corresponds to the case where the standard sensor is irregularly distributed in the light shielding box; The reference database further includes at least one second reference vector, and the processing component is further configured to: Performing gridding processing on the light shielding box based on the light shielding parameters of the light shielding box to obtain at least one grid; Evaluating the credibility of the at least one grid based on the target parameter and the position distribution data of the at least one group of sensors to be tested in the box, wherein the credibility reflects the reliability of the readings of the sensors to be tested in a certain grid; Determining the at least one second reference vector based on the light loss of the at least one grid, the number of the standard sensors, and the reliability; the second reference vector is one of the reference vectors used to determine the spatial distribution of the standard sensors, and the second reference vector corresponds to a situation where the standard sensors are neatly arranged in the light shielding box; determining a reading accuracy of the at least one group of sensors to be tested based on the first reading and the second reading; In response to at least one of the readings meeting the warning condition, a prompt message is issued; wherein, A difference between the number of the standard sensors and the number of the sensors to be tested is greater than a preset value, and the preset value is determined based on the light intensity.

2. The system according to claim 1, wherein The processing component is further configured to: Determining a theoretical reading of the sensor to be tested based on the first reading, setting data of the at least one group of standard sensors, position distribution data of the at least one group of sensors to be tested, and the target parameter; The reading accuracy is determined based on the theoretical reading and the second reading.

3. The system according to claim 2, wherein: The processing component is further configured to: constructing a prediction map based on the first readings, the setting data of the at least one group of standard sensors, the position distribution data of the at least one group of sensors to be tested, and the target parameter, wherein the prediction map reflects the characteristics of the position distribution of the standard sensors, the sensors to be tested, and the adjustable light source in the light shielding box; Based on the prediction map, the theoretical reading is determined by a reading prediction model, and the reading prediction model is a machine learning model.

4. A method for testing an inductive sensor, characterized in that: The method is executed by a processing component of the inductive sensor testing system according to any one of claims 1 to 3, and includes: In response to the adjustable light source being activated according to target parameters, obtaining first readings of at least one set of standard sensors and second readings of at least one set of sensors to be tested; the target parameters including the illumination intensity of the adjustable light source; According to the target parameters and the position distribution data of the at least one group of sensors to be tested in the light shielding box, the setting data of the at least one group of standard sensors are determined by referring to a database, wherein the setting data represents the position distribution of the at least one group of standard sensors in the light shielding box; The reference database includes at least one first reference vector, which is one of the reference vectors used to determine the spatial distribution of the standard sensor, and the first reference vector corresponds to the case where the standard sensor is irregularly distributed in the light shielding box; The reference database further includes at least one second reference vector, and the method further includes: Performing gridding processing on the light shielding box based on the light shielding parameters of the light shielding box to obtain at least one grid; Evaluating the credibility of the at least one grid based on the target parameter and the position distribution data of the at least one group of sensors to be tested in the box, wherein the credibility reflects the reliability of the readings of the sensors to be tested in a certain grid; Determining the at least one second reference vector based on the light loss of the at least one grid, the number of the standard sensors, and the reliability; the second reference vector is one of the reference vectors used to determine the spatial distribution of the standard sensors, and the second reference vector corresponds to a situation where the standard sensors are neatly arranged in the light shielding box; determining a reading accuracy of the at least one group of sensors to be tested based on the first reading and the second reading; In response to at least one of the readings meeting the warning condition, a prompt message is issued; wherein, A difference between the number of the standard sensors and the number of the sensors to be tested is greater than a preset value, and the preset value is determined based on the light intensity.

5. The method according to claim 4, wherein Determining the reading accuracy of the at least one group of sensors to be tested based on the first reading and the second reading includes: Determining a theoretical reading of the sensor to be tested based on the first reading, setting data of the at least one group of standard sensors, position distribution data of the at least one group of sensors to be tested, and the target parameter; The reading accuracy is determined based on the theoretical reading and the second reading.

6. The method according to claim 5, wherein Determining a theoretical reading of the sensor to be tested based on the first reading, setting data of the at least one group of standard sensors, position distribution data of the at least one group of sensors to be tested, and the target parameter includes: Constructing a prediction map based on the first readings, the setting data of the at least one group of standard sensors, the position distribution data of the at least one group of sensors to be tested, and the target parameter, wherein the prediction map reflects the characteristics of the position distribution of the standard sensors, the sensors to be tested, and the adjustable light source in the light shielding box; Based on the prediction map, the theoretical reading is determined by a reading prediction model, and the reading prediction model is a machine learning model.

Images