Error determination method for a sensor, production system, program product and storage medium

Abstract

The application discloses a sensor error determination method, a production system, a program product and a storage medium. The method is based on a preset calibration data set to construct a mapping relationship between a walking distance and a walking time. The calibration data set includes time parameters and distance parameters. The time parameters include timestamp data of at least two marking points on a calibration sample when the at least two marking points pass through at least one calibration station. The distance parameters include physical distance data between the at least two marking points. Based on the timestamp data and the mapping relationship, the calculated distance parameters of the calibration sample when passing through the at least one calibration station are calculated. The calculated distance parameters and read distance parameters are fitted to determine error parameters of the sensor. The read distance parameters include walking readings collected by the sensor when the at least two marking points pass through the at least one calibration station. The calibration and calibration of the sensor error can be realized without stopping and disassembling the equipment, and the calibration process is greatly simplified, and the production interruption and material loss are reduced.

Description

Technical Field

[0001] This application relates to the field of energy storage technology, specifically to a method for determining sensor error, a production system, a computer program product, and a non-transitory computer-readable storage medium. Background Technology

[0002] Continuous roll-to-roll production lines are widely used in industrial manufacturing fields such as lithium-ion battery electrodes, metal foils, thin film materials, photovoltaic materials, paper, and textiles. On this type of production line, the roll material passes through multiple process stations sequentially along the conveyor belt direction. The spatial relationship of each process station along the conveyor belt direction has a significant impact on the configuration of process parameters, the correlation of test results, and the analysis of the production process.

[0003] In actual production, the conveyor belt distance between each process station is usually used as a basic parameter to estimate the time lag of the process response, align the data collected from different stations, and set equipment parameters.

[0004] In related technologies, the actual conveyor belt distance between each station is usually estimated based on the cumulative readings of the roll material position sensor. However, distance estimation based on the cumulative readings of the roll material position sensor is easily affected by factors such as sensor zero-point offset, proportional error, and roll material slippage, resulting in insufficient accuracy and stability of the distance results. In addition, to improve calibration accuracy, it is often necessary to stop the machine for measurement, disassemble the equipment, or conduct multiple trial runs, which increases production downtime and material waste. Summary of the Invention

[0005] This application provides a method for determining sensor error, a production system, a program product, and a storage medium, which can solve at least one of the above-mentioned technical problems.

[0006] On one hand, embodiments of this application provide a method for determining the error of a sensor, including: Based on the pre-set calibration dataset, a mapping relationship between conveyor distance and conveyor time is constructed. The calibration dataset includes time parameters and distance parameters. The time parameters include timestamp data when at least two markers on the calibration sample pass at least one calibration station. The distance parameters include physical distance data between each pair of at least two markers. Based on timestamp data and mapping relationships, calculate the distance parameters when the calibration sample passes through each calibration station; The fitting calculation distance parameter and the reading distance parameter are used to determine the error parameter of the sensor, wherein the reading distance parameter includes the conveyor belt readings collected by the sensor when at least two marker points pass through at least one calibration station.

[0007] On the other hand, embodiments of this application provide a production system, characterized in that it includes: At least one calibrated workstation; The sensor is used to collect conveyor belt readings as at least two marker points on the calibration sample pass through at least one calibration station. A processor, which is used to execute any of the methods described above.

[0008] On the other hand, embodiments of this application provide a computer-readable storage medium storing a computer program adapted for loading by a processor to execute the sensor error determination method as described in any of the above embodiments.

[0009] On the other hand, embodiments of this application provide a computer device, the computer device including a processor and a memory, the memory storing a computer program, the processor executing the sensor error determination method as described in any of the above embodiments by calling the computer program stored in the memory.

[0010] On the other hand, embodiments of this application provide a computer program product, including computer instructions, which, when executed by a processor, implement the sensor error determination method as described in any of the above embodiments.

[0011] The sensor error determination method provided in this application constructs a mapping relationship between conveyor belt distance and conveyor belt time using a preset calibration dataset. By using the actual physical distance of the marked points as a benchmark, it can eliminate the dependence on equipment drawings, manual measurements, and original sensor readings, thereby improving the accuracy and reliability of distance calculation. By substituting the timestamp into the mapping relationship to calculate the actual distance parameters, the actual conveyor belt distance between workstations can be obtained under real production line operation conditions, effectively eliminating calculation deviations caused by coil stretching, slippage, and equipment vibration. Finally, the conveyor belt distance error parameters are determined by directly fitting the actual distance parameters and the read distance parameters, thus realizing the calibration and standardization of sensor errors without stopping the machine to disassemble the equipment, greatly simplifying the calibration process and reducing production interruptions and material losses. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram illustrating an application scenario of a sensor error determination method provided in an embodiment of this application.

[0014] Figure 2This is a flowchart illustrating the error determination method for a sensor provided in an embodiment of this application.

[0015] Figure 3 and Figure 4 This is a schematic diagram illustrating a scenario for the sensor error determination method provided in an embodiment of this application.

[0016] Figure 5 This is a schematic diagram of the error determination device for the sensor provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Below, we will further introduce the background technology of the embodiments of this application.

[0019] Continuous roll-to-roll production lines are widely used in industrial manufacturing fields such as lithium-ion battery electrodes, metal foils, thin film materials, photovoltaic materials, paper, and textiles. On this type of production line, the roll material passes through multiple process stations sequentially along the conveyor belt direction. The spatial relationship of each process station along the conveyor belt direction has a significant impact on the configuration of process parameters, the correlation of test results, and the analysis of the production process.

[0020] In actual production, the conveyor belt distance between each process station is usually used as a basic parameter to estimate the time lag of the process response, align data collected from different stations, and set equipment parameters. However, due to the large length of the production line, the large number of equipment, and the complex layout, the actual conveyor belt distance between each station is difficult to obtain accurately through simple methods.

[0021] In engineering practice, workstation distances are typically estimated through manual measurement, calculation from equipment installation drawings, or based on cumulative readings from roll material position sensors. These methods have limitations: firstly, manual measurement or drawing calculations struggle to reflect the actual conveyor path of the roll material during operation, resulting in significant measurement errors; secondly, distance estimation based on cumulative readings from roll material position sensors is susceptible to factors such as sensor zero-point offset, proportional errors, and roll material slippage, leading to insufficient accuracy and stability of the distance results. Furthermore, improving calibration accuracy often requires machine shutdown for measurement, equipment disassembly, or multiple trial runs, increasing production downtime and material waste.

[0022] Meanwhile, under continuous production conditions, the passage time of markers or features at different workstations is usually recorded manually or automatically. There are differences in the accuracy of time recording and response delay between different workstations, which leads to the accumulation of errors when calculating spatial position based on time information, making it difficult to meet the requirements of high-precision workstation distance calibration.

[0023] Therefore, how to accurately calibrate the actual conveyor belt distance between each process station on a continuous roll material production line using data available during the production process, without relying on complex equipment modifications, long-term shutdowns, or a large number of test pieces, remains a technical problem that urgently needs to be solved in this field.

[0024] In view of this, embodiments of this application provide a method, apparatus, storage medium, production system, and program product for determining sensor errors. Please refer to... Figure 1 , Figure 1 This is an application scenario diagram of a sensor error determination method for a production system provided in this application embodiment. The application scenario provided in this application includes a production system 1000, which may include at least one calibration station, a sensor, and a processor. The sensor is used to collect the conveyor belt readings when at least two marker points on the calibration material pass through at least one calibration station. The production system can be considered as a roll material production line 100, which refers to an integrated production equipment capable of continuous unwinding, conveying, processing, online inspection, and rewinding of roll materials. It typically arranges multiple functional stations sequentially along the conveyor belt direction to complete the continuous conveying and processing of roll materials at a fixed or adjustable speed, undertaking high-precision, high-efficiency batch production tasks in industrial manufacturing. The application scenarios of roll material production lines are wide-ranging, for example, classified by the type of processed materials and processes: (1) Lithium battery electrode production line: Taking the positive and negative electrode sheets of lithium-ion batteries as the processing objects, it completes processes such as coating, rolling, slitting, CCD detection, and winding. It is a link in the lithium battery manufacturing process and has extremely high requirements for station spacing, conveyor belt accuracy and detection synchronization.

[0025] (2) Photovoltaic thin film production line: used for continuous forming, coating, curing and cutting of photovoltaic backsheets, photovoltaic thin films and other materials, emphasizing the stability of long-distance conveyor belt and the spatiotemporal alignment capability of multi-station detection data.

[0026] (3) Metal foil and film production line: for flexible materials such as aluminum foil, copper foil and plastic film, complete the processes of rolling, surface treatment, slitting and defect detection, and have strict requirements for the ability to suppress the slippage and stretching error of the roll material.

[0027] (4) Printing and packaging production line: used for continuous printing, lamination, die-cutting and rewinding of paper, labels and packaging films, relying on precise station distance to achieve registration, positioning and cutting control.

[0028] (5) Textile and paper production line: suitable for continuous drying, embossing, slitting and winding of fabric, paper and other wide-format materials, with long distance, large span and multi-station collaboration as the main features.

[0029] In order to meet the needs of high-precision manufacturing and quality traceability, the roll material production line with multi-station collaboration, real-time data acquisition and belt distance calibration functions has become the mainstream in the industry. The roll material production line adopts a continuous belt conveyor mode and is equipped with roll material position sensors, timing acquisition devices and data processing units. It can complete station distance calibration and sensor calibration under the conditions of no shutdown and less waste.

[0030] In some embodiments, the roll material production line 100 includes at least one calibration station and a sensor. The calibration station includes a reference station 101 and multiple acquisition stations 102. The reference station can serve as the spatial coordinate origin station of the roll material production line and can be set at the roll unwinding position. It serves as a reference benchmark for production line distance calculation and station positioning and is used to determine the relative conveyor belt distance of each acquisition station.

[0031] The acquisition station can be a station arranged sequentially behind the reference station along the direction of the roll material's conveyor belt. For example, the acquisition station may include a pre-processing station, a roll forming station, a laser thickness measurement station, a slitting station, a CCD inspection station, and a winding station, etc. Each station can perform different processing on the roll material.

[0032] Among them, the sensor can be a detection device used to collect and calibrate the cumulative conveyor length of the sample in real time, and can output continuous cumulative displacement readings.

[0033] Optionally, the production system may also include a timing acquisition unit, which may be a time recording device deployed at each workstation to collect the timestamp data of each marker passing through the corresponding workstation, so as to ensure the timing synchronization of data from multiple workstations.

[0034] The roll material production line 100 can establish a data connection with the computing and processing equipment 200 to cooperate with the computing and processing equipment 200 to realize the station distance calibration and sensor error determination method of this application.

[0035] Optionally, the computing processing device 200 includes at least one of a local industrial control terminal and a remote server.

[0036] The local industrial control terminal may include, but is not limited to, industrial computers, industrial control computers, touch screen terminals, embedded controllers, portable industrial control equipment, etc., and this application embodiment does not limit this.

[0037] The server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud computing, cloud data storage, industrial data processing, and online calibration services. This application does not limit this.

[0038] The method for calibrating the station distance and determining the sensor error of the roll material production line in this application can be implemented independently by the local controller of the roll material production line, or it can be implemented collaboratively by the roll material production line and the computing processing equipment, and there is no limitation on this.

[0039] Based on the above-described scenarios, this application provides a method for determining sensor error. The method for determining sensor error will be described in detail below.

[0040] Please see Figure 2 This application provides a method for determining the error of a sensor, which is implemented by steps 011 to 013, as described in detail below.

[0041] Step 011: Based on the preset calibration dataset, construct the mapping relationship between conveyor distance and conveyor time. The calibration dataset includes time parameters and distance parameters. The time parameters include timestamp data when at least two markers on the calibration sample pass at least one calibration station. The distance parameters include physical distance data between each pair of at least two markers.

[0042] The calibration dataset may include time parameters and distance parameters, which can be data collected when the calibration sample moves on the roll production line.

[0043] The conveyor belt distance can be the cumulative physical length of the calibrated sample traveling along the preset conveyor belt direction on the roll material production line, and the unit can be meters or millimeters.

[0044] Among them, the conveyor belt time can be the time consumed for the calibration material to travel from the reference station to each calibration station on the roll material production line, and the unit can be seconds or milliseconds, etc.

[0045] The mapping relationship can be used to describe the relationship between the conveyor time and the conveyor distance. It can be presented in the form of a function (e.g., S=f(t), where S is the conveyor distance and t is the conveyor time, etc.) to realize the conversion of any conveyor time to the corresponding conveyor distance.

[0046] The time parameter can be the time dimension data of the calibration dataset, or the timestamp data of the markers on the calibration sample when they pass through each calibration station.

[0047] The distance parameter can be the spatial dimension data of the calibration dataset, or the actual physical distance between adjacent marker points on the calibration sample. The data is a known fixed value.

[0048] Among them, the calibration sample can be a roll material sample used for calibrating the station distance in the roll material production line and determining the sensor error.

[0049] Please refer to Figure 3 The markers can be discrete markers set on the calibration sample, serving as reference points for collecting timestamp data and physical distance data. The physical distance between adjacent markers is precisely known, providing a spatial distance reference for constructing the mapping relationship.

[0050] Among them, the calibration station can be a designated station selected on the roll material production line for collecting timestamp data of the marker point. It is a spatial reference position for calibration, and includes at least one reference station. Multiple collection stations can be added as needed.

[0051] The timestamp data can be precise time data recorded by the timing acquisition device when the marker point passes through the calibration station, which can reflect the time node when the marker point passes through each station.

[0052] Among them, the physical distance data can be the actual physical length data between two adjacent marker points on the calibration sample. The data is a fixed known value and is not affected by the operating status of the production line. It serves as a spatial reference for constructing the mapping relationship.

[0053] Specifically, a continuous time-space mathematical mapping model can be constructed based on known discrete spatiotemporal reference points to realize the transformation from discrete data to continuous relationships. By using the spatial truth value of the known physical distance of the marker point and the time acquisition value of the marker point through the timestamp of the workstation, a calibrated discrete dataset can be used to fit the continuous correspondence between conveyor belt time and conveyor belt distance.

[0054] In a roll material production line, the calibration material is usually continuously conveyed at a preset speed. The conveyor distance and conveyor time are continuously correlated. The physical distance data and timestamp data of the marker points can be discrete sampling points on this continuous relationship. Moreover, the physical distance data is the true value that is not affected by the operation of the production line. Based on this, a mapping relationship can be constructed, which can truly reflect the inherent correlation between the material conveyor time and conveyor distance under the actual working conditions of the production line. This eliminates the dependence on equipment drawings, manual measurement or original sensor readings, and provides an accurate spatiotemporal conversion benchmark for subsequent calculation of the actual conveyor distance of the calibration station and determination of sensor error parameters.

[0055] Meanwhile, using the known physical distance of the marker point as a fixed benchmark, the timestamp data is bound to the spatial truth value to ensure that the constructed mapping relationship is real and accurate, which can effectively eliminate the spatiotemporal correlation deviation caused by actual working conditions of the production line such as roll material stretching, slippage, and equipment vibration.

[0056] Step 012: Based on timestamp data and mapping relationships, calculate the distance parameters when the calibration sample passes through each calibration station.

[0057] Among them, the calculated distance parameter can be the cumulative conveyor distance value when the calibration material passes through the corresponding calibration station, which is calculated by substituting the timestamp data of the calibration station into the mapping relationship between conveyor distance and conveyor time. It can be a parameter used to characterize the spatial position of the calibration station relative to the reference origin, and serve as a true reference for subsequent calculation of the actual conveyor distance of the station and calibration of sensor errors.

[0058] Specifically, by utilizing the established continuous mapping relationship between conveyor time and conveyor distance, the accurate conversion of timestamp data to spatial distance parameters can be achieved. Essentially, discrete marker point timestamp data is substituted into the mapping function as independent variables, and the cumulative conveyor distance (calculation distance parameter) when the calibration sample passes through each calibration station is obtained by solving the function.

[0059] In continuous roll production lines, the calibration material travels continuously along the conveyor belt at a stable speed. The conveyor belt travel time and distance typically exhibit a continuous correlation. This mapping relationship reflects the inherent pattern of this correlation under actual production line conditions. Furthermore, this mapping relationship is anchored to the known physical distance of the marker point as the spatial truth, possessing high accuracy and reliability. Based on this, the timestamp data of the marker points recorded at each calibration station corresponds to the conveyor belt travel time of the calibration material when it passes through that station. Substituting this data into the mapping relationship allows for the calculation of the distance parameter corresponding to that station.

[0060] Step 013: Fit and calculate the distance parameters and read the distance parameters to determine the sensor's error parameters. The read distance parameters include the conveyor belt readings collected by the sensor when at least two marker points pass through at least one calibration station.

[0061] The reading distance parameter can be the cumulative displacement reading synchronously collected by the roll material position sensor when the marked point passes through each calibration station, which can be obtained by the encoder.

[0062] Among them, the error parameters of the sensor can be parameters used to describe the deviation between the sensor reading and the actual distance, and can include zero-point offset error and proportional coefficient error, which are used to compensate and correct the sensor reading.

[0063] Among them, the roll material position sensor can be a displacement detection device deployed on the conveyor mechanism of the roll material production line, used to output the cumulative conveyor length of the roll material operation.

[0064] Among them, the conveyor belt reading can be the cumulative displacement value output by the sensor at the corresponding moment, which directly reflects the sensor's measurement result of the conveyor belt distance.

[0065] Specifically, the sensor's systematic error can be extracted by using the model-calculated distance as the true benchmark and through numerical fitting. Under the condition of continuous, uniform roll material operation, the actual conveyor distance and sensor readings should exhibit a stable linear deviation relationship. This deviation can be uniformly represented as a linear model including zero-point offset and a proportionality coefficient. By fitting multiple sets of synchronously acquired calculated distance parameters with the reading distance parameters, the optimal coefficients of this linear model, i.e., the sensor's systematic error parameters, can be solved. The entire process is based on the principle of minimizing the deviation between the true and observed values, utilizing a large number of effective data points to offset the influence of random noise, thereby separating the sensor's inherent zero-point offset error and proportionality error, providing an accurate mathematical model for subsequent reading correction.

[0066] For example, multiple sets of true and observed data pairs can be formed by pairing the calculated distance parameter and the read distance parameter corresponding to the same marker point and time, and invalid data points marked during anomaly verification are removed. Then, the least squares method is used to fit these multiple sets of paired data to minimize the sum of squared residuals for all data points, thus obtaining the optimal proportional gain and zero-point offset error. Based on the error parameters obtained from the initial fitting, the read distance parameters are corrected, the residuals are calculated, and hidden anomalies are further removed by combining the median absolute deviation and the station sensitivity coefficient. The cleaned data is then used to refit and update the error parameters. Finally, the proportional gain error and zero-point offset error obtained from the iterative convergence are used together as the final error parameters of the sensor, thus completing the sensor error calculation.

[0067] In other words, firstly, a mapping relationship between conveyor belt distance and conveyor belt time is established. Then, the actual conveyor belt distance (i.e., theoretical conveyor belt distance) is calculated by inputting timestamp data into the mapping relationship. Next, an error fitting relationship is constructed using the actual conveyor belt distance as the true value and the cumulative reading of the sensor as the observed value. The systematic error parameters of the sensor are then solved using methods such as least squares. For example, the systematic error parameters may include proportional coefficient error and zero-point offset error. The systematic error parameters compensate for and calibrate the readings of the roll material position sensor.

[0068] Thus, by constructing a mapping relationship between conveyor belt distance and conveyor belt time using a pre-set calibration dataset, and taking the actual physical distance of the marked points as the benchmark, the reliance on equipment drawings, manual measurements, and original sensor readings can be eliminated, improving the accuracy and reliability of distance calculation. Furthermore, by substituting the timestamp into the mapping relationship to calculate the actual distance parameters, the actual conveyor belt distance between workstations can be obtained under real production line operation conditions, effectively eliminating calculation deviations caused by coil stretching, slippage, and equipment vibration. Finally, the conveyor belt distance error parameters can be determined by directly fitting the actual distance parameters and the read distance parameters, thereby achieving the calibration and standardization of sensor errors without stopping the machine to disassemble the equipment, greatly simplifying the calibration process and reducing production interruptions and material losses.

[0069] In some implementations, at least one calibration station includes a reference station and at least one data acquisition station, and the method further includes: Step 014: Control the calibration sample to start from the reference station and pass through each collection station in sequence at a preset speed threshold; Step 015: Read the timestamp data of each marker point as it passes through each data acquisition station; Step 016: Generate a calibration dataset based on physical distance data and timestamp data.

[0070] Among them, the reference station can be a station selected on the coil production line as the origin of spatial and temporal reference. It is set at the unwinding start position and serves as the reference for the distance calculation of the entire production line. It can be regarded as the reference zero point of the calibration system.

[0071] The data acquisition station can be a process station or inspection station arranged after the reference station along the direction of the roll material conveyor belt. It is used to collect the timestamp of the passing time of the marker point and the sensor conveyor belt reading to provide raw data for the calibration dataset.

[0072] Among them, the standard conveyor speed set by the preset speed threshold production line controls the calibration material to run continuously at a stable speed close to normal production, ensuring that the collected data matches the actual working conditions and avoiding abnormal timing and displacement caused by sudden speed changes.

[0073] Specifically, under uniform and stable conveyor belt conditions, a global reference origin is established through a baseline station. Synchronous timing signals are generated at multiple stations using markers with known spacing. Then, starting from the baseline station, the calibration material is controlled to pass through each acquisition station sequentially at a preset speed threshold, causing the markers to trigger corresponding acquisitions at each station under a uniform motion pattern. The passage time of each marker at different stations reflects the actual travel time of the roll material on the production line; while the known physical distance between markers provides a true spatial scale unaffected by sensors, slippage, or stretching. By matching and integrating the physical distance with the timestamps from multiple stations, a raw dataset that accurately reflects the time-space correspondence of the roll material is formed, providing a reliable, unified, and reproducible data foundation for subsequent modeling, calculation, and error calibration.

[0074] For example, please see Figure 4 The process can be achieved by pre-determining a baseline station and at least one data acquisition station on the production line, confirming station numbering and layout, setting at least two marker points on the calibration sample, measuring and recording the physical distance data between adjacent marker points, and then starting the roll material production line at a preset speed threshold. This means controlling the drive mechanism to operate stably at the preset speed threshold, allowing the calibration sample to pass through all data acquisition stations sequentially along the conveyor path from the baseline station. During the material's conveyor journey, the detection units at each data acquisition station monitor the marker point passing signals in real time. Each time a marker point is detected, the current precise time is immediately recorded, forming a timestamp data for that marker point at that data acquisition station. This timestamp data is then associated and stored according to the marker number and station number. Finally, the timestamp data from all data acquisition stations, the known physical distance data between marker points, and the synchronously acquired sensor conveyor readings are aligned and integrated, structured according to the marker point order and station order, and any obviously missing or duplicate records are removed. This results in a complete and standardized calibration dataset for subsequent mapping relationship construction.

[0075] By setting up markers with known physical spacing on the calibration sample, uniformly numbering the workstations and markers, and simultaneously collecting the timestamp data of the markers passing through the workstations and the sensor conveyor belt readings, a calibration dataset containing multi-dimensional data can be constructed in a standardized manner, providing a complete and unified data foundation for subsequent mapping relationship construction and error determination. At the same time, controlling the continuous operation of the calibration sample at a preset speed allows data to be collected under real production line conditions, ensuring that the calibration results closely match the actual production status.

[0076] In some implementations, step 016: Based on physical distance data and timestamp data, generate a calibration dataset, including: Step 0161: Calculate the time difference between two adjacent markers based on their timestamp data. Step 0162: When the time difference is greater than 0, calculate the acquisition speed parameter between two adjacent markers based on the physical distance data and the time difference between the two adjacent markers; Step 0163: If the acquisition speed parameters meet the preset speed range, determine the marker point as a normal marker point. The preset speed range is determined based on the preset speed threshold (the moving speed of the sample). Step 0164: Summarize the target timestamp data and target physical distance data corresponding to each normal marker point to generate a calibration dataset.

[0077] The time difference between two markers can be the difference in timestamps between the later marker and the earlier marker at the same data acquisition station, reflecting the time interval between the markers' movement on the production line.

[0078] Among them, the acquisition speed parameter can be the instantaneous conveyor speed calculated based on the known physical distance and corresponding time difference between adjacent marker points, which can be used to determine whether the current acquisition data conforms to the normal movement law of the production line.

[0079] The preset speed threshold can be the standard belt speed set during the production line calibration process.

[0080] The preset speed range can be a reasonable speed interval centered on a preset speed threshold and combined with the allowable fluctuation range, used to identify invalid data with abnormal speeds.

[0081] Among them, normal markers can be those that have been verified as valid and reliable through both time and speed reasonableness checks, and their corresponding timestamps and distance data can be used to construct a calibration dataset.

[0082] The target timestamp data can be the filtered valid timestamp data corresponding to normal marker points, which will be used for subsequent modeling calculations.

[0083] Among them, the target physical distance data can be the known and accurate physical distances corresponding to normal marker points, which can be used as the spatial ground truth of the calibration dataset.

[0084] Specifically, when the roll material is running on a continuous production line, the conveyor speed should be maintained within a stable and continuous reasonable range. The time difference between adjacent marker points must be positive; otherwise, the timing logic will be abnormal. The calculated acquisition speed should also fall within the speed range allowed by the process; otherwise, it indicates anomalies such as trigger jitter, signal loss, or recognition errors. Therefore, through a dual screening mechanism of time difference validity verification and acquisition speed reasonableness verification, abnormal data such as timing reversal, signal jumps, false triggers, and speed abrupt changes can be effectively eliminated, ensuring that the data entering subsequent modeling all come from the normal conveyor process, thereby improving the accuracy and robustness of the mapping relationship.

[0085] The time difference between adjacent markers at the same acquisition station can be calculated by subtracting their timestamp data according to their marker numbers. Then, it's checked whether the time difference is greater than 0. If the time difference is less than or equal to 0, it indicates a timing anomaly, data duplication, or acquisition error, and the data can be directly identified as an anomaly and removed. When the time difference is greater than 0, the known physical distance data between adjacent markers is divided by the corresponding time difference to obtain the acquisition speed parameter for that interval. This acquisition speed parameter is compared with a preset speed range based on a preset speed threshold. If the acquisition speed is within the range, it's considered a normal marker; otherwise, it's considered an anomaly and removed. Finally, the target timestamp data and target physical distance data corresponding to all normal markers that have passed dual verification are structured, summarized, and sorted to form a clean and reliable calibration dataset.

[0086] By performing dual verification on the time difference and acquisition speed parameters of the marker points, invalid marker point data with reversed time or abnormal speed are removed. This achieves preliminary cleaning of the calibration dataset, effectively filtering out abnormal data caused by trigger delays or recording errors during the acquisition process, and improving the accuracy and stability of subsequent mapping model construction. (Supplementary manual: When the time difference is less than or equal to 0, the marker point is identified as an abnormal marker point. Abnormal marker points are removed during data processing, and normal marker points are retained to generate the dataset. Data cleaning further improves the model accuracy.)

[0087] In some implementations, step 011: establishing a mapping relationship between transport distance and transport time includes: Step 0111: Interpolate the timestamp data and physical distance data of each marker point to generate a transport mapping model. The transport mapping model is used to calibrate the cumulative transport distance of the sample over time.

[0088] The conveyor mapping model can be a continuous mathematical function between conveyor distance and conveyor time, denoted as S=f(t), and the corresponding cumulative conveyor distance S of the roll material can be calculated based on any time t.

[0089] The timestamp data represents the precise time point at which each marker point passes the calibration station.

[0090] Specifically, interpolation calculations are performed on the timestamp data and physical distance data of each marker point to obtain a continuous time frame. The distance function enables the calculation of the conveyor belt distance at any given time. Under uniform or stable conveyor belt operation, time and cumulative conveyor belt distance show a continuous and smooth correlation. Although the number of marker points is limited, providing only a few discrete moments of true distance, an interpolation algorithm can construct a smooth curve that conforms to the motion law between these points. This allows the system to calculate the true and reliable cumulative conveyor belt distance for any moment where no markers were collected. First, the markers are arranged in numerical order, and the known physical distance data is accumulated sequentially to obtain the cumulative conveyor belt distance corresponding to each marker point. This is then paired with the timestamp of the marker point to form an ordered set of (time, cumulative distance) discrete points. Piecewise linear interpolation or smooth spline interpolation can then be used to ensure the curve is continuous and conforms to the actual conveyor belt movement. For example, by using time as the x-axis and cumulative conveyor belt distance as the y-axis, interpolation fitting is performed on the discrete points to generate a continuous function covering the entire calibration time period, i.e., the conveyor belt mapping model. The conveyor belt mapping model can output the final, directly callable time. The distance mapping relationship allows you to input any valid time and output the corresponding cumulative conveyor distance, which can be used for subsequent workstation distance calculation and sensor error calibration.

[0091] Thus, by interpolating the timestamp data and physical distance data of the marker points to construct a transport mapping model, a continuous correspondence between time and cumulative transport distance can be established, enabling accurate calculation of the transport distance of the calibrated sample at any time. This compensates for the deficiency that discrete marker points cannot cover the distance calculation throughout the entire time period, ensuring the continuity and accuracy of the transport distance calculation.

[0092] In some implementations, step 012: Based on timestamp data and mapping relationships, calculate the calculated distance parameters as the calibration sample passes through each calibration station, including: Step 0121: Input the timestamp data of each target workstation into the conveyor mapping model so that the conveyor mapping model outputs the mapping distance parameters corresponding to each target workstation; Step 0122: Determine the calculation distance parameters based on each mapping distance parameter.

[0093] The target station can be the calibration station where the current conveyor belt distance needs to be calculated, the reference station, or any data collection station set along the conveyor belt direction.

[0094] Among them, the belt transport mapping model can be a time-distance continuous mathematical model generated by interpolation, which is used to calculate the corresponding cumulative belt transport distance of the roll material at any time.

[0095] The mapping distance parameter can be the cumulative transport distance value directly output by the model after a single timestamp is input into the transport mapping model.

[0096] Specifically, by utilizing established continuous spatiotemporal mapping relationships, the time-series data from multiple workstations are uniformly converted into spatial distance data, and statistical methods are used to improve the reliability of the results. The movement of the roll material on the production line is continuous and smooth, and the belt-carrying mapping model accurately represents time. The correspondence between cumulative conveyor belt distances is established. By substituting the timestamp of each marker point recorded at the target workstation into the model, the actual cumulative conveyor belt distance at that moment can be obtained. Since a workstation passes through multiple marker points, multiple mapped distance parameters will be obtained. By fusing these parameters using statistical methods such as mean and median, random errors such as acquisition jitter, timing deviation, and slight slippage can be offset, ultimately yielding stable and accurate calculated distance parameters, which serve as the true benchmark for subsequent workstation spacing calculations and sensor error calibration.

[0097] For example, timestamp data can be grouped by workstation. All timestamps corresponding to the marked points can be grouped according to their respective target workstations, ensuring that the time-series data for each workstation is calculated independently. Then, each valid timestamp under the same target workstation is sequentially input into the conveyor mapping model. The model calculates and outputs the corresponding mapping distance parameters based on the interpolation function, forming a set of distance calculation results for that workstation. Finally, statistical processing is performed on the multiple mapping distance parameters of the current target workstation, such as calculating the arithmetic mean and median. The statistical results are used as the final calculated distance parameters for that target workstation. Simultaneously, the standard deviation can be calculated as needed to evaluate the stability of the workstation's calibration data. The calculation is repeated for the baseline workstation and all collected workstations to obtain the calculated distance parameters for each workstation on the entire production line, providing a unified true value for subsequent relative distance calculations and error fitting.

[0098] By inputting the timestamp data of the target station into the conveyor mapping model, the mapping distance parameters are obtained. Based on multiple sets of mapping distance parameters, the final calculated distance parameters are determined. The influence of single acquisition error can be eliminated through statistical optimization of multiple sets of data, further improving the accuracy and reliability of the calculated distance parameters when the calibration sample passes through the calibration station.

[0099] In some implementations, step 013: fitting and calculating distance parameters and reading distance parameters to determine sensor error parameters includes: Step 0131: Fit and calculate the distance parameters and reading distance parameters to obtain the various reading error parameters of the sensor; Step 0132: Correct the reading distance parameter based on the target reading error parameter, and calculate the residual between the corrected reading distance parameter and the calculated distance parameter. The target reading error parameter is the median or mean of each reading error parameter. Step 0133: Determine the median absolute deviation of the residuals as the global threshold, and use the ratio of the global threshold to the preset sensitivity coefficient of the calibration station as the station threshold. Summarize the time parameters and distance parameters where the residuals are less than the station threshold to update the calibration dataset. Step 0134: After the calibration dataset has been updated, the process of interpolating the timestamp data and physical distance data of each marker point is repeated to generate the transport mapping model and update the transport mapping model. Step 0135: Determine the sensor error parameters based on the difference between the calculated distance parameters and the read distance parameters output by the updated belt mapping model.

[0100] The reading error parameter can be the difference between the calculated distance parameter and the sensor reading distance parameter, which characterizes the deviation between a single sensor reading and the actual distance.

[0101] The target readout error parameter can be an error value obtained statistically from multiple readout error parameters, which can be the median or the mean, and is used to characterize the overall zero-point offset characteristics of the sensor.

[0102] The corrected reading distance parameter can be the distance value obtained by correcting the original sensor reading using the target reading error parameter, thus initially eliminating the systematic offset of the sensor.

[0103] The residual can be the difference between the sensor reading after preliminary correction and the calculated distance parameter, which is used to identify hidden anomalies that still exist in the data.

[0104] Among them, Median Absolute Deviation (MAD) is a robust statistic based on the median, used to calculate the dispersion of residuals, thereby determining the anomaly detection threshold.

[0105] The global threshold can be a unified anomaly detection threshold obtained based on the median absolute deviation of the residuals, used for global filtering of abnormal data.

[0106] The preset sensitivity coefficient can be a weighting coefficient set in advance for different calibrated workstations. It is used to adjust the strictness of the threshold according to the importance of the workstation. The higher the sensitivity, the stricter the threshold.

[0107] Among them, the workstation threshold can be a specific abnormal threshold applicable to the current workstation, obtained by adjusting the global threshold according to the workstation sensitivity coefficient.

[0108] The updated calibration dataset can be a higher quality and cleaner dataset obtained after removing latent outliers with residuals exceeding the threshold.

[0109] The updated transport mapping model can be reconstructed using the cleaned dataset through re-interpolation. The distance mapping model offers higher accuracy and stronger resistance to interference.

[0110] Specifically, the process begins by subtracting the calculated distance parameter from the sensor reading to obtain the initial error. The overall sensor offset characteristics are then extracted using the median or mean, and preliminary correction is performed. Next, a robust threshold is constructed using the median absolute deviation of the residuals, and combined with workstation sensitivity, differentiated and refined anomaly screening is achieved, eliminating hidden anomalies missed in the first data cleaning. The cleaned data is then used to reconstruct a more accurate conveyor belt mapping model, and the more reliable calculated distance parameter is recalculated. Finally, the complete system error parameters of the sensor are fitted to obtain the final result. For example, the difference between the one-to-one calculated distance parameter and the reading distance parameter can be used to obtain the reading error parameter for each set of data, reflecting the deviation between the sensor reading and the true value. The median or mean of all reading error parameters is taken as the target reading error parameter. This parameter is used to correct the original reading distance parameter, resulting in a corrected reading. The residual between the corrected reading and the calculated distance parameter is then calculated, and the median absolute deviation of the residual is used to determine the global threshold. The global threshold is divided by the preset sensitivity coefficient of the current calibration station to obtain the station threshold. Data with residuals less than the station threshold are retained, and outliers are removed to form an updated calibration dataset. Using the new dataset after removing outliers, the timestamps of the marker points and the physical distance data are re-interpolated to generate a more accurate updated conveyor mapping model. The timestamps of each station are substituted into the updated model to obtain more accurate calculated distance parameters. These are then fitted again with the sensor reading distance parameters to determine the final sensor error parameters, including zero-point offset and scaling factor.

[0111] By using residual verification and preset sensitivity coefficients to adaptively eliminate hidden abnormal data, and iteratively updating the calibration dataset and the tape mapping model, it can deeply filter out difficult-to-identify abnormal data such as manual timing deviation and slight sensor drift, thus achieving secondary purification of the dataset. At the same time, by using least squares fitting to determine the sensor's zero-point offset error and proportional coefficient error, it can completely solve the sensor system error, ensuring the comprehensiveness of error determination and calibration accuracy.

[0112] In some implementations, the method further includes: Step 017: Use the calculated distance parameter corresponding to the target calibration station as the benchmark reference value, and determine the conveyor distance of each calibration station based on the difference between the calculated distance parameter of each calibration station and the benchmark reference value. The target calibration station includes the first station through which the calibration sample passes.

[0113] The target calibration station can be the station selected as the reference starting point for the entire distance. In this scheme, it can be the first station through which the calibration sample passes, which is generally the benchmark station.

[0114] The difference calculation involves subtracting the reference value from the calculated distance parameter of each acquisition station, and the result is the actual conveyor belt distance of that station relative to the reference station.

[0115] Specifically, by establishing a unified global spatial benchmark, the absolute cumulative distance of each workstation is converted into the actual workstation spacing relative to the benchmark workstation. The roll material travels continuously on the production line, and the calculated distance parameters for each workstation are cumulative distances starting from the starting point. To obtain the actual spacing between workstations, a unified benchmark needs to be selected, typically the first workstation the roll material reaches as the benchmark reference point. By subtracting the calculated distance parameters of all other workstations from this benchmark reference value, the actual conveyor belt distance of each workstation relative to the benchmark workstation can be directly obtained, thus establishing a unified, accurate, and directly usable spatial coordinate system for the entire production line's process configuration. For example, the first calibration station through which the calibration sample passes is set as the target calibration station, i.e., the reference station. Then, the reference reference value is obtained, and the calculated distance parameter corresponding to the reference station is read. This is used as the globally unified reference reference value. All calibration stations are traversed, and the calculated distance parameters of the reference station and each other acquisition station are obtained in turn. The relative conveyor distance is calculated. For each calibration station, the calculated distance parameter is subtracted from the reference reference value to obtain the conveyor distance of the station relative to the reference station. The station spacing result is output. The relative conveyor distances of all stations are sorted and output to form complete production line station spatial position parameters, which are used for subsequent equipment control, timing alignment, lag compensation, etc.

[0116] Using the average calculated distance parameter of the benchmark station as a unified benchmark reference value, the difference between each calibrated station and the benchmark reference value is calculated to determine the conveyor belt distance of the station. This can establish a unified spatial coordinate benchmark for the entire production line, eliminate systematic errors caused by benchmark deviation, and achieve accurate calculation of the relative distance of each station, providing accurate spatial parameter basis for production line process configuration, data alignment and equipment control.

[0117] To facilitate better implementation of the sensor error determination method of this application embodiment, this application embodiment also provides a sensor error determination device. Please refer to... Figure 5 , Figure 5 This is a schematic diagram of the sensor error determination device provided in an embodiment of this application. The sensor error determination device may include a construction module 201, a calculation module 202, and a fitting module 203, wherein... Module 201 is used to construct a mapping relationship between conveyor distance and conveyor time based on a preset calibration dataset. The calibration dataset includes time parameters and distance parameters. The time parameters include timestamp data when at least two markers on the calibration sample pass at least one calibration station. The distance parameters include physical distance data between each pair of at least two markers. Calculation module 202 is used to calculate the distance parameters of the calibration sample when it passes through each calibration station based on timestamp data and mapping relationship; The fitting module 203 is used to fit and calculate distance parameters and read distance parameters to determine the error parameters of the sensor. The read distance parameters include the conveyor belt readings collected by the sensor when at least two marker points pass through at least one calibration station.

[0118] Each unit in the aforementioned sensor error determination device 300 can be implemented entirely or partially through software, hardware, or a combination thereof. These units can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each unit.

[0119] The sensor error determination device 300 can be integrated into a terminal or server that has storage and a processor and thus computing capabilities, or the sensor error determination device 300 can be the terminal or server.

[0120] Optionally, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0121] This application also provides a computer-readable storage medium for storing a computer program. This computer-readable storage medium can be applied to a computer device, and the computer program causes the computer device to execute the corresponding processes in the sensor error determination method of the embodiments of this application; for the sake of brevity, these will not be elaborated further here.

[0122] This application also provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the corresponding process in the sensor error determination method of the embodiments of this application. For simplicity, further details are omitted here.

[0123] This application also provides a computer program comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the corresponding steps in the sensor error determination method of this application. For brevity, these steps will not be elaborated further here.

[0124] It should be understood that the processor in this application may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor described above can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0125] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0126] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0127] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0129] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0130] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0131] In addition, the functional units in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0132] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer or a server) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0133] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining the error of a sensor, characterized in that, include: Based on a preset calibration dataset, a mapping relationship between conveyor belt distance and conveyor belt time is constructed. The calibration dataset includes time parameters and distance parameters. The time parameters include timestamp data when at least two marker points on the calibration sample pass at least one calibration station. The distance parameters include physical distance data between each of the at least two marker points. Based on the timestamp data and the mapping relationship, the calculation distance parameters of the calibration sample when it passes through each calibration station are calculated; The calculated distance parameter and the read distance parameter are fitted to determine the error parameter of the sensor, wherein the read distance parameter includes the conveyor belt reading collected by the sensor when the at least two marker points pass through the at least one calibration station.

2. The error determination method for a sensor according to claim 1, characterized in that, The at least one calibration station includes a reference station and at least one data acquisition station, and the method further includes: The calibration sample is controlled to start from the reference station and pass through each of the acquisition stations sequentially at a preset speed threshold. Read the timestamp data of each of the marked points as they pass through each of the data collection stations; The calibration dataset is generated based on the physical distance data and the timestamp data.

3. The error determination method for a sensor according to claim 2, characterized in that, The process of generating the calibration dataset based on the physical distance data and the timestamp data includes: Based on the timestamp data of two adjacent marker points, the time difference between two adjacent marker points is calculated sequentially. When the time difference is greater than 0, the acquisition speed parameter between the two adjacent marker points is calculated based on the physical distance data between the two adjacent marker points and the time difference; If the acquisition speed parameter meets the preset speed range, the marker point is determined to be a normal marker point, and the preset speed range is determined based on the preset speed threshold. The target timestamp data and target physical distance data corresponding to each of the normal marker points are aggregated to generate the calibration dataset.

4. The error determination method for a sensor according to any one of claims 1-3, characterized in that, The mapping relationship between the conveyor belt distance and the conveyor belt time is established, including: Interpolation calculations are performed on the timestamp data and physical distance data of each of the marked points to generate a transport mapping model, which is used to calculate the cumulative transport distance of the calibrated sample over time.

5. The error determination method for a sensor according to claim 4, characterized in that, The calculation of the distance parameters for the calibrated sample as it passes through each of the calibration stations, based on the timestamp data and the mapping relationship, includes: The timestamp data of each target workstation is input into the conveyor mapping model so that the conveyor mapping model outputs the mapping distance parameters corresponding to each target workstation. The calculated distance parameters are determined based on each of the mapping distance parameters.

6. The error determination method for a sensor according to claim 5, characterized in that, The fitting of the calculated distance parameters and the read distance parameters to determine the sensor's error parameters includes: The calculated distance parameters and the read distance parameters are fitted to obtain the various read error parameters of the sensor; The reading distance parameter is corrected based on the target reading error parameter, and the residual between the corrected reading distance parameter and the calculated distance parameter is calculated. The target reading error parameter is the median or mean of each of the reading error parameters. The median absolute deviation of the residuals is determined as the global threshold, and the ratio of the global threshold to the preset sensitivity coefficient of the calibration station is used as the station threshold. The time parameters and distance parameters where the residuals are less than the station threshold are summarized to update the calibration dataset. Once the calibration dataset update is complete, the process of interpolating the timestamp data and physical distance data of each marker point to generate a transport mapping model is repeated to update the transport mapping model. The error parameters of the sensor are determined based on the difference between the calculated distance parameters output by the updated belt mapping model and the readout distance parameters.

7. The method for determining the error of a sensor according to any one of claims 1-6, characterized in that, Also includes: The calculated distance parameter corresponding to the target calibration station is used as the benchmark reference value, and the conveyor distance of each calibration station is determined according to the difference between the calculated distance parameter of each calibration station and the benchmark reference value. The target calibration station includes the first station through which the calibration sample passes.

8. A production system, characterized in that, include: At least one calibrated workstation; The sensor is used to collect conveyor belt readings as the at least two marker points on the calibration sample pass through the at least one calibration station; A processor for performing the method according to any one of claims 1-7.

9. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the error determination method for the sensor according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.

Images