Fault identification processing method and device based on internet of things digital weighing system

Abstract

The application discloses a kind of based on the fault identification processing method and device of Internet of Things digital weighing system, the method includes: fault detection step: track and record the weighing data and state of each way sensor of Internet of Things digital weighing system, whether sensor appears fault is judged;Output substitution step: if sensor appears fault, according to the historical weighing data of the sensor and scale weighing gravity center coordinates generate predicted weighing data, predicted weighing data is used as the output of the sensor of this fault, and the weighing value of whole scale is output in combination with the weighing data of other normal sensor.This application can quickly identify and early warning, and effectively eliminate or reduce the influence of sensor abnormal data by learning and using historical weighing data and scale weighing gravity center change rule, effectively improve the reliability of Internet of Things digital weighing system, reduce the production loss such as shutdown and shutdown caused by weighing system sensor abnormal problem of field user.

Description

TECHNICAL FIELD

[0001] The present application relates to the weighing technology field, and particularly to a fault identification processing method and device of a digital weighing system based on the Internet of Things. BACKGROUND

[0002] A conventional weighing system is composed of a weighing carrier (including a scale table, a hopper or a tank, etc.), a weighing sensor (hereinafter referred to as a sensor), a weighing instrument, and external devices (a printer, a display screen, a PLC, etc.). Common sensors include analog sensors and digital sensors. The analog sensor is mainly a strain pressure sensor, which is composed of an elastic body, a strain gauge, a compensation circuit, a cable, etc. The digital sensor is embedded with an A / D processing module (including a power supply circuit, an A / D circuit, a filter circuit, an MCU processor, a communication interface circuit, etc.) on the basis of the analog sensor, directly converts the internal analog electrical signal into a specific digital signal output, and common output interfaces include RS485, RS422, and CAN bus, etc.

[0003] The scale table of the weighing system will generate a scale table corner error due to the following factors, which is manifested in that the same weight placed at different positions of the scale table, especially at the corner area of the scale table, will output different weighing results by the whole scale before corner error calibration, resulting in weighing error: (1) the parameters of each weighing sensor installed on the scale table are inconsistent, including sensor sensitivity and output impedance difference, which leads to uneven signal output and causes corner error; (2) the installation of the scale table and the scale table foundation, such as uneven foundation, loose limiting device or foundation settlement, which may affect the accuracy of the scale table weighing; (3) mechanical deformation of the scale table, especially in the traditional ground balance scale table, due to long-term partial load or overload, the scale table is not rigid enough and is deformed, which affects the force balance of the sensor and further causes the scale table corner error.

[0004] Ideally, the center of gravity of the measured object should fall within the geometric center area formed by the support points of each sensor. If the center of gravity deviates too much, it will cause uneven force on each point sensor, and the nonlinear error of the sensor output itself will further exacerbate the generation of the scale table corner error.

[0005] The weighing instruments and weighing sensors in common weighing systems have the following connection types: (1) Single-channel analog signal acquisition: One or more analog sensors are connected through a sensor junction box or directly connected in parallel on the internal interface of the instrument to merge into one analog signal and then transmitted to the single-channel A / D circuit of the instrument for acquisition and conversion processing; (2) Multi-channel analog signal acquisition: Multiple analog sensors are connected in groups through multiple sensor junction boxes or directly connected to multiple A / D acquisition channels of the instrument for multi-channel conversion and processing, and the weight data of each A / D channel can be obtained independently; (3) Multiple digital signal acquisition: Multiple digital sensors are connected to the instrument through a digital bus (RS485 / RS422 / CAN, etc.), and the instrument can identify the weight data of each digital sensor.

[0006] The Internet of Things (IoT) digital weighing system includes a weighing platform, a combination of weighing sensors, an IoT digital weighing instrument, external devices, and a cloud server, etc. (see reference) Figure 1 The weighing sensor assembly consists of multiple digital sensors or multiple analog sensors.

[0007] The above weighing systems face several technical challenges that need to be overcome:

[0008] When a few (e.g., 1-2) sensors in a weighing system malfunction (e.g., accidental damage to the communication interface of a digital sensor preventing data acquisition, abnormal power supply or A / D circuit of the digital sensor, short circuit or open circuit in the wiring of the analog sensor, damage to the strain gauge inside the analog sensor, etc.), the weighing data from these sensors may not be acquired or may show significant anomalies. In this case, when the weighing instrument processes the overall weighing data, the missing or significantly deviated local weight data may cause the overall weight to be unable to be output or to be significantly abnormal. The weighing instrument may display abnormal weighing data or alarm messages (e.g., weighing data out of tolerance, sensor malfunction, sensor connection abnormality, etc.). When such severely distorted weighing data is used as the basis for condition determination in industrial process control, it will lead to disruption of the weighing-related industrial process control or severe distortion of statistical data. When the weighing instrument displays alarm information, the equipment connected to the instrument will display abnormal information and suspend operation, requiring on-site technicians to troubleshoot the sensor malfunctions one by one. The on-site wiring of load cells is usually quite complex. The sensor wires are mixed with signal lines and power lines of various industrial equipment and are distributed in various cable trays, making troubleshooting difficult. When such problems occur, on-site users usually cannot solve them in a short time and cannot use the weighing system normally, facing the risk of downtime and production stoppage, which has a significant impact on production efficiency.

[0009] In field use, when a few (e.g., 1-2) sensors malfunction unexpectedly, how to enable the weighing system to quickly identify and warn of the fault, and automatically eliminate or reduce the impact of abnormal sensor data, so as to continue to provide highly accurate weighing data, is a technical challenge that needs to be overcome in the design of weighing systems.

[0010] In a weighing system, the probability of two or more sensors malfunctioning simultaneously is very small. Even if it does occur, it is considered a serious system failure that requires timely intervention from technicians to analyze and resolve the issue, ensuring the continued stability of the weighing system. Summary of the Invention

[0011] The technical problem to be solved by the embodiments of the present invention is to provide a fault identification and processing method and device based on an Internet of Things digital weighing system, so as to improve the reliability of the weighing system.

[0012] To address the aforementioned technical problems, this invention proposes a fault identification and processing method for an Internet of Things (IoT) digital weighing system, wherein the weighing system includes multiple sensors, and the method includes:

[0013] Fault detection steps: Track and record the weighing data and status of each sensor in the IoT digital weighing system to determine whether the sensor is malfunctioning;

[0014] Output substitution steps: If a sensor malfunctions, predictive weighing data is generated based on the sensor's historical weighing data and the weighing center coordinates of the scale platform. The predicted weighing data is used as the output of the malfunctioning sensor, and the total weighing value is output by combining the weighing data of other normal sensors.

[0015] Accordingly, embodiments of the present invention also provide a fault identification and processing device based on an Internet of Things (IoT) digital weighing system, comprising:

[0016] Fault detection unit: tracks and records the weighing data and status of each sensor in the IoT digital weighing system to determine whether the sensor has malfunctioned;

[0017] Output substitution unit: If a sensor malfunctions, predictive weighing data is generated based on the sensor's historical weighing data and the weighing center coordinates of the scale platform. The predicted weighing data is used as the output of the malfunctioning sensor, and the total weighing value is output by combining the weighing data of other normal sensors.

[0018] The beneficial effects of this invention are as follows: This invention enables IoT digital weighing systems to quickly identify and provide early warnings of faults in certain weighing sensors on weighing platforms, hoppers, or tanks. By learning and utilizing historical weighing data and patterns of the center of gravity, it effectively eliminates or reduces the impact of abnormal sensor data. Through dynamic redundancy technology, it achieves seamless switching between automatic fault exit and takeover weighing modes, continuing to provide highly accurate weighing data. This effectively improves the reliability of the weighing system and reduces production losses such as downtime caused by sensor malfunctions. Furthermore, this invention supports a weighing algorithm recovery mechanism after sensor faults are resolved, reducing the complexity of system debugging and maintenance, and further enhancing the reliability of IoT digital weighing systems. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of the Internet of Things digital weighing system according to an embodiment of the present invention.

[0020] Figure 2 This is a flowchart illustrating the fault identification and processing method of an IoT-based digital weighing system according to an embodiment of the present invention.

[0021] Figure 3 This is a three-segment interval measurement curve diagram according to an embodiment of the present invention. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] In this embodiment of the invention, directional indicators (such as up, down, left, right, front, back, etc.) are only used to explain the relative positional relationship and movement of each component in a specific posture (as shown in the figure). If the specific posture changes, the directional indicator will also change accordingly.

[0024] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features.

[0025] Please refer to Figure 1The IoT digital weighing system of this invention includes a weighing platform, a weighing sensor assembly, an IoT digital weighing instrument, external devices, and a cloud server. The weighing sensor assembly involves two modes (one is selected for actual use). Mode 1 consists of multiple digital sensors connected to the digital interface of the weighing instrument via a digital communication bus, allowing the instrument to collect multiple channels of digital weight data. Mode 2 consists of multiple analog sensors, with the analog signals from each sensor connected to the multi-channel analog interface of the weighing instrument. The weight data for each A / D channel is independently obtained through multi-channel A / D acquisition and conversion within the instrument. The digital signal communication in Mode 1 and the analog signal communication in Mode 2 can also be converted via a sensor junction box before being connected to the instrument interface. The weighing instrument provides a human-machine interface, parameter setting, total weight calculation, report recording, and communication control processing. External devices interact with the instrument, providing specific functions such as printing, display, and control. External devices mainly include printers, external displays, PLCs, and motors, which can be selected according to the user's actual needs. Weighing instrument data can interact with the cloud server via wireless communication (4G / 5G / Wifi / Bluetooth, etc.) or wired communication (standard Ethernet port or industrial bus, etc.) to achieve functions such as data synchronization management, cloud storage, and remote equipment control. The cloud server can connect to external devices by parsing the instrument's weighing data or control commands, directly enabling remote device control. Remote clients, including mobile apps, web pages, or host computers, can interact with the cloud server to achieve remote control of the weighing instrument or external equipment.

[0026] Please refer to Figure 2 The fault identification and processing method of the Internet of Things-based digital weighing system in this embodiment of the invention includes a fault detection step, an output substitution step, and a fault recovery step.

[0027] Fault detection steps: Track and record the weighing data and status of each sensor in the IoT digital weighing system to determine if the sensor is faulty. The fault detection steps are implemented through the following steps to track and record weighing data and status, and to diagnose faults.

[0028] 1. The output of each sensor is compensated by pre-adjusting the angle difference and the angle difference correction, as shown in Formula 1.

[0029] ;(Formula 1)

[0030] Where W = measured value of the object being measured, n = sensor position number, and i = i-th measurement. =Angular difference coefficient of the nth sensor, W n [i] represents the i-th measurement value of the n-th sensor.

[0031] When adjusting the angular difference, the value of W is fixed. For example, when a 20 kg weight is placed above the No. 1 position sensor, W = 20 kg, and the first set of sensor data is collected. Then, when the weight is successively placed above the other sensors on the weighing platform in turn, another n - 1 sets of sensor data are collected. , , , and the angular difference coefficient is solved according to Formula 1. .

[0032] 2. Due to the non - linear error of the sensor, it is necessary to divide the weight range and use a multi - segment linear function to fit and correct the error, as shown in Formula 2.

[0033] ; (Formula 2)

[0034] Among them, = the actual weight of the measured object, j = the weight range segment, = the slope of the j - th weight range, = the intercept of the j - th weight range.

[0035] Taking the weight calibration of three intervals as an example (see Figure 3 ). The starting point and the ending point of the first weight interval are the loading points of the weights of 2 weights. The actual weights of the loaded weights are known quantities, and the output weight W of the weighing platform of the IoT digital weighing system is the measured value. According to the two - point weight values substituted into Formula 2, and can be obtained. Similarly, and of the second and third weight intervals, and ]> are obtained.

[0036] After weight calibration, according to the weight range segment corresponding to the measured value W of the measured object, the actual weight of the measured object is obtained by using Formula . .

[0037] 3. Calculate the center of gravity point (X0, Y0) of the empty weighing platform and the center of gravity point (X i , Y i ) of the loaded weighing platform (including the measured object).

[0038] The sensors supporting the weighing platform of the IoT digital weighing system usually have common installation distributions such as an equilateral triangle layout of 3 sensors, a rectangular layout of 4 sensors, etc. In the conventional weighing operation, the center of gravity of the measured object is placed in the geometric center area of the weighing platform, such as the geometric center area of an equilateral triangle or a rectangle, to ensure the measurement accuracy. The center of gravity point of the loaded weighing platform (including the measured object) often coincides with or is very close to the center of gravity point of the empty weighing platform.

[0039] Calculate the center of gravity (X0, Y0) of the empty weighing platform and the center of gravity (X0, Y0) of the loaded weighing platform (including the object being measured). i Y i (e.g., formula 3-4).

[0040] ;(Formula 3)

[0041] ;(Formula 4)

[0042] Where i = the i-th measurement, (X i ,Y i ) = the centroid coordinates of the i-th measurement, (A n B n = Coordinates of the installation location of the nth sensor =Measurement value of the nth sensor when the scale is empty = The nth sensor measurement value during the i-th measurement.

[0043] Each sensor is installed at coordinates (A) n B n Once the IoT digital weighing system is designed and installed, the measured value becomes a known fixed value, which can be measured using drawings or a measuring tape. During measurement, each sensor measures... Measurement values ​​of empty scale The weight output of each sensor when the scale is empty is a known quantity. Formula 3 can be used to calculate the center of gravity (X0, Y0) of the empty weighing platform, and Formula 4 can be used to calculate the center of gravity (X0, Y0) of the loaded weighing platform (including the object being measured). i Y i ).

[0044] The center of gravity (X) of the loading platform (including the object being measured). i Y i The deviation between the center of gravity (X0, Y0) of the empty scale platform and the center of gravity of the empty scale platform is calculated according to Formula 5.

[0045] ;(Formula 5)

[0046] Among them, the smaller the value of the m coefficient, the higher the degree of overlap of the center of gravity, and the higher the accuracy of the weighing of the entire scale of the Internet of Things digital weighing system.

[0047] Based on the actual conditions of the scale's commissioning, select an appropriate m coefficient. When the center of gravity deviates from the range of m values ​​(which is set by the user according to the actual situation), the overall weighing accuracy of the IoT digital weighing system is within the user's acceptable range, and this data can be recorded as normal weighing data. For example, when 0 < m < 5%, the overall weighing accuracy of the IoT digital weighing system (e.g., better than 0.03%) will meet the requirements of the on-site user.

[0048] 4. Record the weighing data and center of gravity status for each measurement process (i-th weighing) at regular intervals, including the actual weight of the object being measured in the i-th weighing. Weight output of each sensor The center of gravity (X) of the loading platform (including the object being measured) i Y i The system records each normal weighing data point (where the center of gravity deviation corresponds to a coefficient within the set m value range). Weighing data from an empty scale is recorded in the same format. Heartbeat monitoring is used to record weight data periodically (the timing value is configurable, such as recording every 5 seconds). The weighing data storage structure is as follows:

[0049] .

[0050] Weighing data in the empty state is recorded in the same format, as follows:

[0051] .

[0052] The default weight of the empty scale is 0, or 0 after a zeroing operation. The center of gravity of the empty scale platform is (X0, Y0). The weight output of each sensor when the scale is empty is... .

[0053] 5. When the weight output of the weighing sensor is abnormal, the instrument of the IoT digital weighing system will immediately identify the faulty sensor.

[0054] When a few (e.g., 1-2) load cells in an IoT digital weighing system malfunction, such as due to accidental damage to the digital sensor communication interface leading to the inability to collect weighing data, abnormalities in the internal power supply or A / D circuit of the digital sensor, short circuits or open circuits in the analog sensor wiring, or damage to the internal strain gauge of the analog sensor, the weighing data from these sensors will be unavailable or will show significant anomalies. Since each load cell weighs independently, the instrument can identify the fault condition and time in real time. If no sensor data is collected or the load cell data is significantly abnormal (the difference between the current data and the previous weighing data exceeds the threshold set by the instrument), a warning will be issued, indicating the faulty sensor number and abnormal state, reminding the user to address the issue.

[0055] This invention employs a pre-adjusted weighing platform angle difference and angle difference correction algorithm to compensate for the output of each sensor, and uses a multi-segment linear function to fit and correct the measurement error of the entire weighing platform, thereby improving the overall measurement accuracy. This invention also uses a heartbeat monitoring method to automatically track and record the weighing data of each sensor, as well as the center of gravity coordinates of the weighing platform in both empty and loaded states. When the weighing sensor outputs abnormally, the fault is automatically identified and located by comparing it with historical weighing data.

[0056] Output substitution step: If a sensor malfunctions, predicted weighing data is generated based on the sensor's historical weighing data and the scale's center of gravity coordinates. This predicted weighing data is used as the output of the malfunctioning sensor (ensuring the entire scale functions normally and the accuracy of the weighing data is within the user's acceptable range). The total scale weighing value is then output by combining this predicted data with weighing data from other functioning sensors. The output substitution step involves the following steps to perform the calculations for weighing data prediction and fault substitution.

[0057] 1. Predict the weighing data of faulty sensors.

[0058] When only one load cell fails, the current weighing data of the failed load cell is predicted using the following two methods:

[0059] Method a. Direct selection or One of them is used as the predicted value;

[0060] Where h is the location number of the faulty sensor. The weight value of sensor h (i+1th weighing) is derived from the X-coordinate of the center of gravity. It is the (i+1)th weighing value of sensor h, derived from the Y-coordinate of the center of gravity.

[0061] ;

[0062] ;

[0063] Method b. Using a weighted algorithm, weighting coefficients are set based on the distribution of sensors on the weighing platform, and the predicted value is calculated using the following formula:

[0064] ;

[0065] in These are weighting coefficients. ;

[0066] When two sensors malfunction, the current predicted values ​​of both sensors are treated as unknowns. Combined with the current weight output values ​​of other functioning sensors and the previous center of gravity coordinates, the predicted values ​​of the two sensors can be solved by constructing the following system of equations:

[0067] .

[0068] Taking a faulty sensor (h=1) as an example, the instrument retrieves the last normal weighing data at the time the fault occurred, for example, the weighing data of the i-th time, to predict the weighing data of the (i+1)-th time. The weighing data of the i-th time is known to be... During the (i+1)th weighing, the center of gravity of the weighing platform (including the object being measured) remains unchanged, still being [the center of gravity]. ,Right now , In the (i+1)th weighing, only sensor number 1 needs to predict data. In addition, the weight output of other normal sensors is .

[0069] Based on formula 4, the following is derived:

[0070] ;

[0071] ;

[0072] Thus, we obtain formulas 6 and 7:

[0073] ;(Formula 6)

[0074] ;(Formula 7)

[0075] in, The weight value of sensor 1 at the (i+1)th weighing is derived from the X-coordinate of the center of gravity. It is the (i+1)th weighing value of sensor 1 derived from the Y-coordinate of the center of gravity.

[0076] The predicted value of the (i+1)th weighing by sensor #1 can be calculated in the following two ways:

[0077] Method a. Directly select one of the two options as the predicted value;

[0078] Method b. Employing a weighted algorithm, the weighting coefficients are flexibly set based on the distribution of sensors on the weighing platform, and the predicted value is calculated using Formula 8. This method can effectively improve the accuracy of the predicted value.

[0079] ;(Formula 8)

[0080] in These are weighting coefficients. . Initial value = 0.5. Further evaluate the position of the center of gravity of the weighing platform (including the object being measured) from the center of gravity of the empty weighing platform. If the X coordinate deviates more than the Y coordinate, then q is less than 0.5; if the Y coordinate deviates more than the X coordinate, then q is greater than 0.5.

[0081] When two load cells malfunction (taking load cells 1 and 2 as an example), construct the following system of equations:

[0082] ;

[0083] Where i = the i-th measurement, (X i ,Y i ) = the centroid coordinates of the i-th measurement, (A n B n = Coordinates of the installation location of the nth sensor =Measurement value of the nth sensor when the scale is empty = The measurement value of the nth sensor during the (i+1)th measurement.

[0084] The equations only contain the predicted value of sensor number 1. And the predicted value of sensor No. 2 Since it is an unknown quantity, it can be solved by a system of equations.

[0085] 2. Based on the predicted value of the faulty sensor and the weighing data of other normal sensors, calculate the output weight of the entire scale according to Formula 2. And continue to record the weighing data after this compensation and repair. The system will collect and monitor subsequent weighing data, while continuing to collect and monitor the weight output data of the faulty sensor. The coordinates of the center of gravity of the weighing platform (including the object being measured) will remain unchanged until the sensor fault is resolved.

[0086] This invention employs digital redundancy technology, utilizing the center-of-gravity coordinates and output data from other normal sensors in historical normal weighing data to generate predicted data for faulty sensors. This predicted data replaces the output of the faulty sensor, ensuring normal weighing operation of the entire scale and keeping the accuracy of the weighing data within acceptable ranges for the user. When only one weighing sensor malfunctions, the predicted value of that faulty sensor can be independently calculated using the X and Y coordinates of the scale's center-of-gravity. Based on the sensor distribution on the scale, the weighting coefficients are flexibly adjusted, and a weighted algorithm is used to calculate the predicted value of the faulty sensor, improving prediction accuracy.

[0087] Fault recovery steps: Stop predicting sensor weighing data according to user settings; or when the IoT digital weighing system is empty, compare the collected sensor weighing data, total scale weight measurement value, and center of gravity coordinates with the normal weighing data in the empty scale state. If the difference is lower than the preset threshold and the center of gravity coordinate deviation is within the preset range, stop prediction, output the normal weight value, and update the normal weighing data and coordinate data.

[0088] When the weighing system is in the weight compensation and repair state of a faulty sensor, the instrument will continuously collect and monitor the weight output of the faulty sensor. Once the fault is cleared, the sensor output will return to normal. The instrument supports both automatic and manual modes to exit the weight compensation and repair state.

[0089] 1. After troubleshooting the sensor malfunction, technicians can manually operate the instrument to exit the weight compensation repair state and restore normal weighing.

[0090] 2. When the IoT digital weighing system is empty, the instrument can automatically determine the sensor fault recovery status. If the conditions are met, it will automatically exit the weight compensation and repair state and resume normal weighing.

[0091] When the scale is empty, the IoT digital weighing system compares the collected sensor data, the total weight measurement, and the center of gravity coordinates with the normal weighing data for an empty scale. The normal weighing data for an empty scale is... When the difference between the normal sensor weight and the normal output weight in the empty scale state is lower than the threshold range set by the instrument, and the m coefficient corresponding to the center of gravity coordinate deviation value (see Formula 5) is within the range of m values ​​set by the instrument, the instrument will automatically exit the weight compensation and repair algorithm, continue to output the weight according to the normal weighing algorithm, and update the normal weighing data and coordinate data.

[0092] This invention continuously monitors the weighing system status in real time and supports a weighing algorithm recovery mechanism after sensor faults are resolved. When a sensor fault is resolved, the weight compensation repair algorithm can be exited manually via instrument operation, or it can be automatically exited by comparing historical weight data from each sensor and the center-of-gravity coordinates of the empty scale with historical data from the sensors and the empty scale. This improves the fault tolerance and reliability of the weighing system while ensuring the accuracy of the overall scale measurement data. Seamless switching between automatic fault exit and takeover weighing modes is achieved, effectively reducing on-site maintenance costs.

[0093] The fault identification and processing device based on the Internet of Things digital weighing system in this embodiment of the invention includes a fault detection unit, an output substitution unit, and a fault recovery unit.

[0094] Fault Detection Unit: Tracks and records the weighing data and status of each sensor in the IoT digital weighing system to determine if a sensor has malfunctioned. This invention employs digital redundancy technology, using the center-of-gravity coordinates from historical weighing data to generate predictive data for faulty sensors, replacing their output and ensuring normal weighing operation of the entire scale. The accuracy of the weighing data is controlled within acceptable ranges for the user. Seamless switching between automatic fault exit and takeover weighing modes is achieved.

[0095] Output substitution unit: If a sensor malfunctions, predictive weighing data is generated based on the sensor's historical weighing data and the weighing center coordinates of the scale platform. The predicted weighing data is used as the output of the malfunctioning sensor, and the total weighing value is output by combining the weighing data of other normal sensors.

[0096] As one implementation method, the fault detection unit uses pre-adjusted angle difference and angle difference correction to compensate for the output of each sensor, and uses a multi-segment linear function to fit and correct the measurement error of the IoT digital weighing system.

[0097] As one implementation method, the fault detection unit compensates for the sensor output according to the following formula: ;

[0098] Where W is the measured value of the object being measured, n is the position number of the sensor, and i represents the i-th measurement. W is the angular difference coefficient of the nth sensor. n [i] represents the i-th measurement value from the n-th sensor;

[0099] The error is corrected by fitting the following formula:

[0100] ;

[0101] in, The actual weight of the object being measured is given by , and j represents the weight range. This represents the slope of the j-th weight interval. Let be the intercept of the j-th weight interval;

[0102] The center of gravity of the empty weighing platform (X0, Y0) and the center of gravity of the loaded weighing platform (X0, Y0) are calculated according to the following formulas. i Y i ):

[0103] ;

[0104] ;

[0105] Among them, (X) i Y i (A) represents the centroid coordinates of the i-th measurement. n B n () represents the coordinates of the installation location of the nth sensor. This is the measurement value of the nth sensor when the scale is empty. This represents the measurement value of the nth sensor during the i-th measurement;

[0106] The center of gravity (X) of the loading platform is calculated using the following formula. i Y i The deviation between the center of gravity (X0, Y0) of the empty scale platform and the center of gravity of the scale platform:

[0107] ;

[0108] m is a preset coefficient.

[0109] As one implementation method, the fault detection unit uses heartbeat monitoring to periodically track and record the weighing data of each sensor, the coordinate data of the center of gravity of the weighing platform in both empty and loaded states. When the sensor's weight output fails, the fault is identified and located by comparing it with historical weighing data.

[0110] In one implementation, when only one sensor malfunctions, the output replacement unit retrieves the last normal weighing data at the time of the malfunction and predicts the current weighing data of the malfunctioning sensor based on the following two methods:

[0111] Method a. Direct selection or One of them is used as the predicted value;

[0112] Where h is the location number of the faulty sensor. The weight value of sensor h (i+1th weighing) is derived from the X-coordinate of the center of gravity. It is the (i+1)th weighing value of sensor h, derived from the Y-coordinate of the center of gravity.

[0113] ;

[0114] ;

[0115] Method b. Using a weighted algorithm, weighting coefficients are set based on the distribution of sensors on the weighing platform, and the predicted value is calculated using the following formula:

[0116] ;

[0117] in These are weighting coefficients. ;

[0118] When two sensors malfunction, the predicted values ​​of the two sensors are treated as unknowns. By combining the current weight output value of other normal sensors with the previous center of gravity coordinates, the predicted values ​​of the two sensors can be solved by constructing the following system of equations.

[0119] ;

[0120] When only one weighing sensor malfunctions, this invention can independently calculate the predicted value of the faulty sensor using the X and Y coordinates of the weighing platform's center of gravity. Based on the distribution of sensors on the weighing platform, the weighting coefficients can be flexibly adjusted, and the predicted value of the faulty sensor can be calculated using a weighted algorithm, thereby improving the prediction accuracy.

[0121] In one implementation, the output replacement unit calculates the output weight of the entire scale according to the following formula, based on the predicted value of the faulty sensor and the weighing data from other normal sensors. We will continue to record the weighing data after this compensation and repair, as well as subsequent weighing data, while continuing to collect and monitor the weight output data of the faulty sensor:

[0122] .

[0123] The fault recovery unit stops predicting the sensor's weighing data according to the user's settings; or when the IoT digital weighing system is empty, it compares the collected sensor weighing data, the total weight measurement value, and the center of gravity coordinates with the normal weighing data in the empty state. If the difference is lower than the preset threshold and the center of gravity coordinate deviation is within the preset range, it stops predicting, outputs the normal weight value, and updates the normal weighing data and coordinate data.

[0124] This invention continuously monitors the weighing system status in real time and supports a weighing algorithm recovery mechanism after sensor faults are resolved. When a sensor fault is resolved, the weight compensation repair algorithm can be exited manually via instrument operation, or it can be automatically exited by comparing historical weight data from each sensor and the center-of-gravity coordinates of the empty scale with historical data from the sensors. This improves the fault tolerance and reliability of the weighing system while ensuring the accuracy of the overall scale measurement data. This invention achieves seamless switching between automatic fault exit and takeover weighing modes, effectively reducing on-site maintenance costs.

[0125] When certain weighing sensors in the weighing platform, hopper, or tank of an Internet of Things (IoT) digital weighing system malfunction, the present invention can quickly identify and issue early warnings, and effectively eliminate or reduce the impact of abnormal sensor data by learning and utilizing historical weighing data and the changing patterns of the weighing center of gravity of the weighing platform.

[0126] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A fault identification and processing method for an Internet of Things (IoT)-based digital weighing system, wherein the weighing system includes multiple sensors, characterized in that, The method includes: Fault detection steps: Track and record the weighing data and status of each sensor in the IoT digital weighing system to determine whether the sensor is malfunctioning; Output substitution steps: If a sensor malfunctions, predictive weighing data is generated based on the sensor's historical weighing data and the weighing center coordinates of the scale platform. The predicted weighing data is used as the output of the malfunctioning sensor, and the total weighing value is output by combining the weighing data of other normal sensors. In the fault detection step, pre-adjustment angle difference and angle difference correction are used to compensate for the output of each sensor, and multi-segment linear functions are used to fit and correct the measurement error of the IoT digital weighing system. In the fault detection process, the sensor output is compensated according to the following formula: ； Where W is the measured value of the object being measured, n is the position number of the sensor, and i represents the i-th measurement. W is the angular difference coefficient of the nth sensor. n [i] represents the i-th measurement value from the n-th sensor; The error is corrected by fitting the following formula: ； in, The actual weight of the object being measured is given by , and j represents the weight range. This represents the slope of the j-th weight interval. Let be the intercept of the j-th weight interval; The center of gravity of the empty weighing platform (X0, Y0) and the center of gravity of the loaded weighing platform (X0, Y0) are calculated according to the following formulas. i Y i ): ； ； Among them, (X) i Y i (A) represents the centroid coordinates of the i-th measurement. n B n () represents the coordinates of the installation location of the nth sensor. This is the measurement value of the nth sensor when the scale is empty. This represents the measurement value of the nth sensor during the i-th measurement; The center of gravity (X) of the loading platform is calculated using the following formula. i Y i The deviation between the center of gravity (X0, Y0) of the empty scale platform and the center of gravity of the scale platform: ； m is a preset coefficient; In the output replacement step, when a sensor malfunctions, the last normal weighing data at the time of the malfunction is retrieved, and the current weighing data of the malfunctioning sensor is predicted according to the following two methods: Method a. Direct selection or One of them is used as the predicted value; Where h is the location number of the faulty sensor. The weight value of sensor h (i+1th weighing) is derived from the X-coordinate of the center of gravity. It is the (i+1)th weighing value of sensor h, derived from the Y-coordinate of the center of gravity. ； ； Method b. Using a weighted algorithm, weighting coefficients are set based on the distribution of sensors on the weighing platform, and the predicted value is calculated using the following formula: ； in These are weighting coefficients. .

2. The fault identification and processing method for an IoT-based digital weighing system as described in claim 1, characterized in that, In the fault detection process, a heartbeat monitoring method is used to track and record the weighing data of each sensor, the coordinate data of the center of gravity of the weighing platform in both empty and loaded states at regular intervals. When the sensor's weight output fails, the fault is identified and located by comparing it with historical weighing data.

3. The fault identification and processing method for an IoT-based digital weighing system as described in claim 1, characterized in that, The output replacement step also includes a fault recovery step: stop predicting the sensor's weighing data according to the user's settings; or when the IoT digital weighing system is in an empty state, compare the collected sensor weighing data, the total weight measurement value, and the center of gravity coordinates with the normal weighing data in an empty state. If the difference is lower than a preset threshold and the center of gravity coordinate deviation is within a preset range, then stop the prediction, output the normal weight value, and update the normal weighing data and coordinate data.

4. A fault identification and processing device based on an Internet of Things (IoT) digital weighing system, characterized in that, include: Fault detection unit: tracks and records the weighing data and status of each sensor in the IoT digital weighing system to determine whether the sensor has malfunctioned; Output replacement unit: If a sensor malfunctions, predictive weighing data is generated based on the sensor's historical weighing data and the weighing center coordinates of the scale platform. The predicted weighing data is used as the output of the malfunctioning sensor, and the total weighing value is output by combining the weighing data of other normal sensors. The fault detection unit compensates for the sensor output according to the following formula: ； Where W is the measured value of the object being measured, n is the position number of the sensor, and i represents the i-th measurement. W is the angular difference coefficient of the nth sensor. n [i] represents the i-th measurement value from the n-th sensor; The error is corrected by fitting the following formula: ； in, The actual weight of the object being measured is given by , and j represents the weight range. This represents the slope of the j-th weight interval. Let be the intercept of the j-th weight interval; The center of gravity of the empty weighing platform (X0, Y0) and the center of gravity of the loaded weighing platform (X0, Y0) are calculated according to the following formulas. i Y i ): ； ； Among them, (X) i ,Y i (A) represents the centroid coordinates of the i-th measurement. n B n () represents the coordinates of the installation location of the nth sensor. This is the measurement value of the nth sensor when the scale is empty. This represents the measurement value of the nth sensor during the i-th measurement; The center of gravity (X) of the loading platform is calculated using the following formula. i Y i The deviation between the center of gravity (X0, Y0) of the empty scale platform and the center of gravity of the scale platform: ； m is a preset coefficient; When only one sensor fails, the output replacement unit retrieves the last normal weighing data at the time of the failure and predicts the current weighing data of the failed sensor based on the following two methods: Method a. Direct selection or One of them is used as the predicted value; Where h is the location number of the faulty sensor. The weight value of sensor h (i+1th weighing) is derived from the X-coordinate of the center of gravity. It is the (i+1)th weighing value of sensor h, derived from the Y-coordinate of the center of gravity. ； ； Method b. Using a weighted algorithm, weighting coefficients are set based on the distribution of sensors on the weighing platform, and the predicted value is calculated using the following formula: ； in These are weighting coefficients. .

5. The fault identification and processing device for an IoT-based digital weighing system as described in claim 4, characterized in that, The fault detection unit uses pre-adjusted angle difference and angle difference correction to compensate for the output of each sensor, and uses a multi-segment linear function to fit and correct the measurement error of the IoT digital weighing system.

Images