Electromagnetic force weighing dynamic calibration method and system

Abstract

The invention provides an electromagnetic force weighing dynamic calibration method and system, and relates to the technical field of metering testing, and the method comprises the steps: 1, collecting a composite vibration signal for the high-speed movement, frequent start and stop and random vibration states of a package in logistics sorting, and forming working condition vibration characteristic data; step 2, based on the working condition vibration characteristic data, constructing a random vibration excitation signal covering a full frequency band so as to generate a signal for simulating an excitation environment of a multi-frequency superposition electromagnetic force balancing device in an actual working condition; and step 3, based on the random vibration excitation signal, in the calibration process of the electromagnetic force weighing sensor, synchronously acquiring an electromagnetic force signal output by the sensor, a real-time acceleration signal of the measured object and vibration spectrum data, and recording amplitude-frequency characteristic change conditions of different vibration components on the electromagnetic force. According to the invention, weighing precision and stability are improved, errors are reduced, and dynamic balance between electromagnetic force and measured mass is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of metrological testing, in particular to an electromagnetic force weighing dynamic calibration method and system. BACKGROUND

[0002] In the logistics sorting, the traditional electromagnetic force weighing calibration technology has some limitations. The simulation of the traditional technology to the vibration environment is not comprehensive enough to cover the actual vibration situation of multi-frequency superposition in the logistics sorting process. For example, when the parcels are conveyed at high speed on the sorting line, the fixed frequency vibration generated by the motor operation of the conveying belt, the random high frequency vibration generated by the collision between the parcels and the transient impact vibration caused by the start and stop of the equipment will exist at the same time. However, the traditional calibration method may only focus on the compensation of the fixed frequency vibration of the motor, and the response to the high frequency vibration generated by the collision is insufficient, so that in the actual weighing, these vibration components which are not fully considered will interfere with the balance relationship between the electromagnetic force and the measured mass, affecting the weighing precision.

[0003] In addition, the compensation mechanism of the traditional technology lacks dynamic adaptability and is difficult to flexibly adjust according to the real-time vibration characteristics. For example, in different periods of the sorting line, the parcel flow is different. When the parcels are dense, the vibration frequency spectrum of the conveying belt will change significantly, and the distribution ratio of the vibration energy in each frequency band is quite different from that when the parcels are sparse. Therefore, under the condition that the parcel flow fluctuates greatly, the weighing error is easy to exceed the preset range, and it is difficult to stably maintain high weighing precision. SUMMARY

[0004] The technical problem to be solved by the present application is to provide an electromagnetic force weighing dynamic calibration method and system, which improves the precision and stability of electromagnetic force weighing through multi-dimensional signal acquisition, dynamic compensation mechanism and parameter adjustment.

[0005] To solve the above technical problems, the technical solutions of the present application are as follows:

[0006] In a first aspect, an electromagnetic force weighing dynamic calibration method is provided, which comprises:

[0007] Step 1. In the logistics sorting, the composite vibration signal is collected for the high-speed motion of the parcels, the frequent start and stop and the random vibration state, and the working condition vibration characteristic data is formed.

[0008] Step 2. Based on the working condition vibration characteristic data, a random vibration excitation signal covering the full frequency band is constructed to generate a signal simulating the excitation environment of the electromagnetic force balance device in the actual working condition with multi-frequency superposition.

[0009] Step 3. Based on the random vibration excitation signal, the electromagnetic force signal output by the sensor, the real-time acceleration signal of the measured object and the vibration frequency spectrum data are synchronously collected in the calibration process of the electromagnetic force weighing sensor, and the amplitude-frequency characteristic change of the electromagnetic force under different vibration components is recorded.

[0010] Step 4, according to the amplitude-frequency characteristic change of electromagnetic force, extract the nonlinear distortion parameters corresponding to each vibration frequency band, and form a comprehensive compensation mechanism;

[0011] Step 5, set three detection points with fixed spatial position on the logistics sorting line, and combine the comprehensive compensation mechanism to obtain the vibration characteristics of each detection point, and generate dynamic compensation coefficients based on the spatial distribution relationship of the three detection points;

[0012] Step 6, according to the real-time vibration characteristics and dynamic compensation coefficients in logistics sorting, dynamically adjust the matching relationship of each frequency band parameter in the comprehensive compensation mechanism;

[0013] Step 7, based on the adjusted comprehensive compensation mechanism, adjust the current of the electromagnetic coil to realize the dynamic balance of the electromagnetic force and the measured mass, and obtain the weighing accuracy under different working conditions.

[0014] Further, based on the working condition vibration characteristic data, a random vibration excitation signal covering the full frequency band is constructed to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition, including:

[0015] Perform time-frequency domain joint analysis on the collected working condition vibration characteristic data, identify the dominant vibration frequency components, energy distribution characteristics of each frequency component, and the change rule of vibration energy with time under the conditions of high-speed motion, frequent start-stop and random vibration of packages in the logistics sorting process;

[0016] Based on the dominant vibration frequency components, combined with the energy distribution characteristics of each frequency component, calculate the energy proportion relationship of each frequency component in the actual working condition, and generate the basic harmonic component;

[0017] In the basic harmonic component, superimpose a random noise component consistent with the statistical characteristics of logistics sorting vibration, adjust the amplitude of the random noise component, and realize the energy matching of the total energy level with the energy of the actual working condition vibration characteristic data through energy normalization processing;

[0018] Based on the adjusted random vibration excitation signal, a wideband random vibration time-varying waveform covering the full working frequency band of the logistics sorting equipment is constructed to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition.

[0019] Further, based on the random vibration excitation signal, in the calibration process of the electromagnetic force weighing sensor, the sensor output electromagnetic force signal, real-time acceleration signal and vibration spectrum data of the measured object are synchronously collected, the amplitude-frequency characteristic change of electromagnetic force under different vibration components is recorded, including:

[0020] Load a random vibration excitation signal to the electromagnetic force balancing device, collect three-dimensional vibration data of the measured object in real time by using an acceleration sensor, and collect the current signal of the electromagnetic coil in real time by using a current detection unit;

[0021] Based on the three-dimensional vibration data, the vibration frequency spectrum data of the measured object is calculated;

[0022] Based on the real-time current signal of the electromagnetic coil and the physical parameters of the electromagnetic force weighing sensor, the electromagnetic force signal is calculated;

[0023] The vibration frequency spectrum data and the electromagnetic force signal are subjected to time-frequency correlation analysis, and the amplitude variation characteristics of the electromagnetic force signal under different vibration frequency components are identified to obtain the correlation analysis result;

[0024] Based on the correlation analysis result, a mapping relationship table of vibration frequency and electromagnetic force amplitude response is established, and the change of the vibration frequency component to the electromagnetic force amplitude in the full frequency band is recorded.

[0025] Further, according to the amplitude-frequency characteristic change of the electromagnetic force, the nonlinear distortion parameters corresponding to each vibration frequency band are extracted to form a comprehensive compensation mechanism, including:

[0026] Based on the vibration frequency-electromagnetic force amplitude response mapping table, the gain variation of the full frequency band electromagnetic force is analyzed, and the abnormal vibration frequency band deviating from the linear theoretical value is obtained through preset gain deviation threshold identification;

[0027] For the abnormal vibration frequency band, the dynamic deviation amount of the electromagnetic force theoretical prediction value and the sensor measured value in the frequency band is calculated, and a frequency-indexed nonlinear distortion parameter set is generated;

[0028] The frequency-indexed nonlinear distortion parameter set is divided according to the abnormal frequency band boundary frequency division rule to establish a segmented frequency domain compensation rule set, forming a dynamic force real-time compensation mechanism.

[0029] Further, three detection points with fixed spatial positions are set on the logistics sorting line, and the vibration characteristics of each detection point are obtained by combining the comprehensive compensation mechanism, and a dynamic compensation coefficient is generated based on the spatial distribution relationship of the three detection points, including:

[0030] In the bearing direction of the weighing station conveyor belt, three vibration sensors are linearly arranged as detection points along the running path of the conveyor belt at a preset fixed interval, the first detection point is located at the starting section of the conveyor belt, the second detection point is located at the middle section of the conveyor belt, and the third detection point is located at the end section of the conveyor belt, and the frequency spectrum distribution data of the acceleration signal of the conveyor belt at each detection point is collected in real time;

[0031] The spectrum distribution data of the three detection points is combined with the corresponding spatial coordinate values, and the rigidity vibration component data and the flexible deformation component data of the conveying belt are obtained according to a predefined vibration component separation rule;

[0032] Based on the rigidity vibration component data and the flexible deformation component data, the dynamic influence ratio at the weighing position is calculated, and an adaptive compensation coefficient set bound to the spatial position is generated.

[0033] Further, according to the real-time vibration characteristics and the dynamic compensation coefficient in the logistics sorting, the matching relationship of the frequency band parameters in the comprehensive compensation mechanism is dynamically adjusted, including:

[0034] The real-time collected vibration spectrum data is input into the segmented frequency domain compensation unit, and the dynamic compensation coefficient set is called to reconstruct the compensation weight parameters of each frequency band;

[0035] Based on the reconstructed compensation weight parameters, the proportion of high-frequency vibration energy in the total spectrum is calculated, and a high-frequency compensation intensity enhancement coefficient is generated according to the proportion value;

[0036] The low-frequency amplitude offset in the compensation weight parameter is extracted, and the high-frequency compensation intensity enhancement coefficient is combined to dynamically correct the frequency band parameters in the compensation unit.

[0037] Further, based on the adjusted comprehensive compensation mechanism, the electromagnetic coil current is adjusted to realize the dynamic balance of the electromagnetic force and the measured mass, and the weighing accuracy under different working conditions is obtained, including:

[0038] The updated comprehensive compensation mechanism parameter set is input into the electromagnetic force control unit, and the real-time collected vibration spectrum data is combined to analyze the vibration disturbance force component, and a current compensation signal group is generated according to the frequency band discrete distribution;

[0039] The compensation amount of each frequency band in the current compensation signal group is extracted, and the amplitude-phase synchronous superposition of the frequency band is performed with the electromagnetic coil reference driving current to generate an anti-vibration synthesized driving current instruction set;

[0040] The electromagnetic coil current is controlled by the anti-vibration synthesized driving current instruction set, and the displacement state sequence of the weighing unit is collected by using the displacement sensing device;

[0041] The displacement state sequence is dynamically detected for stationarity, and when the displacement fluctuation amplitude is within the zeroing tolerance range for a continuous preset period, the mass measurement value after vibration isolation is obtained.

[0042] The second aspect is an electromagnetic force weighing dynamic calibration system, including:

[0043] The signal acquisition module is used to acquire the composite vibration signal and form the working condition vibration characteristic data in the logistics sorting according to the high-speed motion, frequent start-stop and random vibration state of the package.

[0044] A signal construction module is configured to construct a random vibration excitation signal covering a full frequency band based on the vibration characteristic data of the working condition, so as to generate a signal simulating a multi-frequency superimposed electromagnetic force balance device excitation environment in the actual working condition.

[0045] A synchronous acquisition module is configured to synchronously acquire an electromagnetic force signal output by the sensor, a real-time acceleration signal of the measured object and vibration frequency spectrum data in a calibration process of the electromagnetic force weighing sensor based on the random vibration excitation signal, and record amplitude-frequency characteristic changes of the electromagnetic force caused by different vibration components.

[0046] A parameter extraction module is configured to extract nonlinear distortion parameters corresponding to each vibration frequency band according to the amplitude-frequency characteristic changes of the electromagnetic force, and form a comprehensive compensation mechanism.

[0047] A coefficient generation module is configured to set three detection points with fixed spatial positions on the logistics sorting line, acquire vibration characteristics of each detection point in combination with the comprehensive compensation mechanism, generate dynamic compensation coefficients based on spatial distribution relationships of the three detection points.

[0048] A dynamic adjustment module is configured to dynamically adjust matching relationships of frequency band parameters in the comprehensive compensation mechanism according to real-time vibration characteristics in the logistics sorting and the dynamic compensation coefficients.

[0049] A balance adjustment module is configured to adjust a current of the electromagnetic coil based on the adjusted comprehensive compensation mechanism, so as to realize dynamic balance between the electromagnetic force and the measured mass, and obtain weighing precision under different working conditions.

[0050] In a third aspect, a computing device includes:

[0051] one or more processors;

[0052] a storage device storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0053] In a fourth aspect, a computer readable storage medium stores a program, when the program is executed by a processor, the method is implemented.

[0054] The above scheme of the present application at least has the following beneficial effects:

[0055] In the technical layer, the limitations of traditional static calibration or single frequency band vibration compensation are broken through, the random vibration excitation signal covering the whole frequency band is constructed, the complex vibration environment of multi-frequency superposition in the logistics sorting is accurately reproduced, the calibration process is highly consistent with the actual working condition, and a scientific environment simulation basis is provided for improving the weighing precision; in the improvement of the weighing precision, the influence of different vibration components on the amplitude-frequency characteristics of the electromagnetic force is accurately captured by means of time-frequency domain joint analysis and multi-parameter synchronous acquisition, the vibration interference of each frequency band is offset by combining the nonlinear distortion parameter extraction and the segmented frequency domain compensation mechanism, and the weighing accuracy in the complex dynamic scene is improved, for example, in the sorting line with high-speed package shuttle and frequent start-stop, the weighing error can still be controlled in a very low range.

[0056] In terms of adaptability and stability, the dynamic compensation coefficient is generated through the spatial distribution design of the three fixed detection points, and the compensation parameter matching relationship is dynamically adjusted according to the real-time vibration characteristics, which can quickly respond to the working condition changes, ensure that the electromagnetic force and the measured mass are always in dynamic balance, enhance the working stability in the complex environment of random vibration and transient impact, and reduce the weighing data drift caused by vibration fluctuation; in practical application, the weighing reliability of logistics sorting is improved, the sorting errors and delivery delays caused by inaccurate weighing are reduced, and the labor and time cost losses are indirectly reduced. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 is a flowchart of an electromagnetic force weighing dynamic calibration method provided by an embodiment of the application.

[0058] Figure 2 is a schematic diagram of an electromagnetic force weighing dynamic calibration system provided by an embodiment of the application. DETAILED DESCRIPTION

[0059] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0060] As Figure 1 shown, an embodiment of the application proposes an electromagnetic force weighing dynamic calibration method, which comprises the following steps:

[0061] Step 1, in the logistics sorting, for the high-speed motion of the package, the frequent start-stop and the random vibration state, the composite vibration signal is collected to form the working condition vibration characteristic data;

[0062] Step 2, based on the vibration characteristic data of the working condition, a random vibration excitation signal covering the full frequency band is constructed to generate a signal simulating the excitation environment of the electromagnetic force balance device in the actual working condition with multi-frequency superposition;

[0063] Step 3, based on the random vibration excitation signal, during the calibration process of the electromagnetic force weighing sensor, the electromagnetic force signal output by the sensor, the real-time acceleration signal of the measured object and the vibration frequency spectrum data are synchronously collected, and the amplitude-frequency characteristic changes of the electromagnetic force under different vibration components are recorded;

[0064] Step 4, according to the amplitude-frequency characteristic changes of the electromagnetic force, the nonlinear distortion parameters corresponding to each vibration frequency band are extracted to form a comprehensive compensation mechanism;

[0065] Step 5, three detection points with fixed spatial positions are set on the logistics sorting line, and the vibration characteristics of each detection point are obtained combined with the comprehensive compensation mechanism, and based on the spatial distribution relationship of the three detection points, a dynamic compensation coefficient is generated;

[0066] Step 6, according to the real-time vibration characteristics and the dynamic compensation coefficient in the logistics sorting, the matching relationship of the parameters of each frequency band in the comprehensive compensation mechanism is dynamically adjusted;

[0067] Step 7, based on the adjusted comprehensive compensation mechanism, the current of the electromagnetic coil is adjusted to realize the dynamic balance between the electromagnetic force and the measured mass, and the weighing precision under different working conditions is obtained.

[0068] In the embodiment of the present application, the composite vibration signal is collected to form working condition vibration characteristic data, which can accurately capture the real state of the package under high-speed motion, frequent start-stop and random vibration, and ensure that the entire weighing adjustment process is closely matched with the actual working condition from the source; the random vibration excitation signal covering the full frequency band is constructed, which can realistically simulate the complex environment of multi-frequency superposition in the actual working condition, so that the calibration environment of the electromagnetic force weighing sensor is closer to the real sorting scene, reduces the calibration error caused by incomplete environment simulation, and improves the pertinence and effectiveness of calibration; the electromagnetic force signal, real-time acceleration signal and vibration frequency spectrum data are collected synchronously during the calibration process, and the influence of vibration components is recorded, so that the action law of different vibration factors on the electromagnetic force amplitude-frequency characteristic can be comprehensively mastered, the nonlinear distortion parameters are extracted to form a comprehensive compensation mechanism, which can offset the weighing deviation caused by each vibration frequency band, weaken the interference of vibration on the measurement result, reduce the influence of nonlinear distortion on weight measurement, and preliminarily improve the weighing accuracy; three fixed detection points are set on the sorting line, and dynamic compensation coefficients are generated, which can adapt to the vibration differences of different positions in combination with the spatial distribution relationship, so that the compensation mechanism has spatial adaptability, and the weighing precision is not affected by the vibration interference difference caused by the different positions of the detection points; the frequency band parameter matching relationship of the comprehensive compensation mechanism is dynamically adjusted according to the real-time vibration characteristics, so that the compensation mechanism can flexibly respond to the real-time changes of the working condition in the sorting process, enhance the adaptability to complex and variable vibration environment, and adjust the electromagnetic coil current based on the adjusted comprehensive compensation mechanism to realize dynamic balance, which can stably guarantee the accurate matching between the electromagnetic force and the measured mass under various working conditions, improve the weighing precision of logistics sorting, and ensure the reliability of package weight measurement under different motion states, thereby providing strong support for sorting efficiency and accuracy.

[0069] In a preferred embodiment of the present application, the step 1 in the logistics sorting can include:

[0070] In the embodiment of the present application, the selection and deployment of the vibration acquisition equipment are performed, the vibration sensor with wide frequency response capability is selected, and the response frequency range thereof needs to cover all vibration frequencies that can occur in the logistics sorting, from low-frequency vibration generated by friction between the package and the conveying belt during high-speed motion of the package to high-frequency impact vibration in the moment of frequent start-stop; when deployed, the sensor array is uniformly installed on the surface of the conveying belt, so that the detection range of each sensor can cover the motion trajectories of at least two adjacent packages; at the same time, sensors are additionally installed near the motor, gear and other power components of the sorting machine to capture the vibration interference generated by the equipment itself, and all sensors are connected to the data acquisition terminal through shielded cables to reduce electromagnetic interference in the signal transmission process.

[0071] According to the running speed of the logistics sorting line (usually 1-3 meters per second), the sampling frequency of the sensor is set to capture the vibration changes of the package in the minimum movement unit (such as every 1 centimeter movement), for example, when the sorting line speed is 2 meters / second, the sampling interval is not more than 5 milliseconds, ensuring that the instantaneous vibration of the package due to bumps and collisions will not be missed; the collection duration is set to at least cover the complete cycle of 3 consecutive packages from entering the detection area to leaving according to the shortest package spacing of the sorting line (such as 0.5 meters) and the running speed, for example, when the speed is 2 meters / second and the spacing is 0.5 meters, the single collection duration is not less than 2.5 seconds to include the initial vibration when the package enters, the stable vibration in the detection area, and the ending vibration when it leaves.

[0072] When the package enters the sorting line, the sensor starts to capture the vibration signal in real time. In the high-speed motion stage, the friction between the package and the conveyor belt will produce continuous vibration, and the sensor can record the stable waveform of this vibration, including the waveform fluctuations caused by each slight bump. When the sorting machine performs start-stop operation, the package will have a short relative displacement with the conveyor belt due to inertia, at which time the sensor can capture the sudden impact vibration and clearly distinguish the waveform difference between the forward impact when starting and the backward impact when stopping. At the same time, the vibrations in the surrounding environment (such as the operation of adjacent sorting line equipment, ground-conducted vibration) and the vibrations of the sorting equipment itself (such as the periodic vibration of the motor running, the intermittent vibration of the gear meshing) will mix in, forming a random vibration signal superimposed on the package motion vibration. The sensor will convert all these vibration signals of different sources and different natures into a continuous electrical signal stream and transmit it to the data acquisition terminal in real time for temporary storage.

[0073] After data storage, the first level of signal purification processing is physical interference filtering, which removes abnormal peak signals (such as signal values suddenly appearing far beyond the normal range) caused by single sensor failure by comparing the signals of adjacent sensors in the sensor array. The second level is environmental interference stripping, which uses the inherent vibration signal of the equipment near the power components as a reference to subtract this part of regular vibration (such as the 50Hz power frequency vibration of the motor) from the sensor signals in the package detection area, leaving only the vibration component generated by the package motion. The third level is state segment division, which combines the running log of the sorting line (such as the start-stop time of the conveyor belt, the package position record) to accurately cut the continuous vibration signal into independent signal streams of individual packages, and then divide the corresponding motion state sub-signal segment in each package signal stream according to the speed change curve (acceleration segment, constant speed segment, deceleration segment), ensuring that each sub-signal segment strictly corresponds to one motion state of the package.

[0074] For each sub-signal segment, multi-dimensional feature extraction is carried out. In the high-speed uniformity sub-segment, the average amplitude of the vibration signal (reflecting the overall intensity of the vibration), the amplitude standard deviation (reflecting the stability of the vibration), and the number of amplitude peak values in 10 seconds (reflecting the frequency of the bump) are calculated; in the start-up acceleration sub-segment, the time from the baseline to the peak value of the signal (reflecting the impact speed), the difference between the peak amplitude and the baseline amplitude (reflecting the impact strength), and the number of cycles of the signal decay to a stable value after the peak value (reflecting the vibration dissipation speed) are recorded; in the stop deceleration sub-segment, the negative peak value of the signal (due to reverse inertia) and the fluctuation recovery time (from the stop impact to the vibration smoothness) are extracted; for the random vibration component, the energy proportion (the ratio of the vibration energy in a certain frequency interval to the total energy) in different frequency intervals (such as 10-50Hz, 50-200Hz, and above 200Hz) is calculated to determine the influence weight of each frequency band random vibration.

[0075] An independent vibration feature file is established for each package, which contains the package ID, the time of entering the sorting line, the start and end time of each motion state sub-segment, and all extracted feature values corresponding to each sub-segment, such as average amplitude, peak value parameter, and frequency distribution; at the same time, the feature files of multiple packages are summarized to form a vibration feature database covering different weights (from light small items to heavy packages), different materials (hard cartons, soft packages, and metal containers), and different motion speeds, and the specific working condition parameters (such as the conveying belt speed at that time, the environmental temperature, and the equipment running time) corresponding to each feature value are also marked in the database, and finally a working condition vibration feature data is formed which can comprehensively reflect the vibration law of various packages in complex motion state in the logistics sorting scene.

[0076] In a preferred embodiment of the present application, the step 2 of constructing a random vibration excitation signal covering the full frequency band based on the working condition vibration feature data to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balance device in the actual working condition can include:

[0077] Step 220, time-frequency domain joint analysis is performed on the collected working condition vibration feature data to identify the dominant vibration frequency component, the energy distribution characteristics of each frequency component, and the change rule of vibration energy with time in the high-speed motion, frequent start-stop, and random vibration state of the package in the logistics sorting process;

[0078] Step 221, based on the dominant vibration frequency component, the energy distribution characteristics of each frequency component are combined to calculate the energy proportion relationship of each frequency component in the actual working condition, and a basic harmonic component is generated;

[0079] Step 222, in the base harmonic component, superimpose a random noise component consistent with the statistical characteristics of the logistics sorting vibration, adjust the amplitude of the random noise component, and realize the energy matching of the total energy level and the actual working condition vibration characteristic data through energy normalization processing;

[0080] Step 223, based on the adjusted random vibration excitation signal, construct a wideband random vibration time-varying waveform covering the full working frequency band of the logistics sorting equipment, to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balance device in the actual working condition.

[0081] In the embodiment of the application, the collected working condition vibration characteristic data is split into several continuous short period signal segments according to the time axis, the time length of each segment is set to be able to capture a complete vibration period (such as 0.1 seconds), and the time-frequency domain conversion is performed on each segment to obtain the frequency composition corresponding to each time and the energy value of each frequency, forming an original data set containing "time-frequency-energy" three-dimensional information; the cuckoo algorithm is introduced to identify the dominant vibration frequency component, all possible frequency values in the data set are regarded as "bird nests", and the "mass" of each bird nest is determined by the total energy proportion of the frequency in all periods. In the initial stage of the algorithm, a plurality of frequency values are randomly selected as initial "bird nest positions" (i.e. candidate dominant frequencies), the cuckoo laying behavior is simulated, the total energy proportion of each candidate frequency is calculated, the top 30% of "high-quality bird nests" are retained, and the remaining 70% of "inferior bird nests" are updated - by randomly generating new frequency values near the existing high-quality frequencies (such as generating new frequencies of 98-102 Hz near the 100 Hz high-quality frequency), the process of cuckoo finding better habitat is simulated.

[0082] The above process is iterated for multiple times (such as 20 rounds), and finally the frequency value whose total energy proportion accumulates more than 80% is selected as the "dominant vibration frequency component", and the energy values of these dominant frequencies in each period are counted, and the energy-time curve is drawn, so as to determine the "energy distribution characteristics of each frequency component" (such as high energy in the starting stage and low energy in the uniform speed stage) and the "vibration energy-time variation law" (such as the overall energy appearing peak value at the start-stop moment).

[0083] Based on the dominant vibration frequency components determined in step 220, the total energy of each dominant frequency in the entire time period is first calculated (the energy values of the frequency in each time period are accumulated). Then, the total energy of each dominant frequency is divided by the sum of the total energies of all dominant frequencies to obtain the "energy proportion" of each frequency, i.e. the "energy proportion relationship". Based on these proportion relationships, the cuckoo algorithm is introduced again to optimize the harmonic parameters. The harmonic amplitude and initial phase corresponding to each dominant frequency are regarded as "bird nest parameters", and the error value between the synthesized signal of the basic harmonic component and the actual vibration signal is used as the standard to judge the quality of the bird nest (the smaller the error, the higher the quality of the bird nest). In the initial stage, each harmonic is assigned an initial amplitude according to the energy proportion relationship (for example, for a frequency with an energy proportion of 30%, the initial amplitude is set to 30% of the maximum amplitude of the actual signal).

[0084] When the algorithm runs, the harmonic parameters with errors exceeding the threshold value are adjusted. The 20% parameter combinations with the smallest errors are retained, and the remaining parameter combinations are updated in a "Levy flight" manner. The existing high-quality parameters are used as a reference, and the amplitude (e.g. increased or decreased by 5%) and initial phase (e.g. increased or decreased by 5°) are randomly adjusted in a small range. This simulates the behavior of cuckoos searching for new nests at a long distance. After multiple iterations (e.g. 30 rounds), the error between the synthesized signal and the actual signal is less than 1%. At this time, the harmonic combination is the "basic harmonic component".

[0085] The vibration signals of the logistics sorting line running continuously for 24 hours are collected, and the chaotic vibration components of non-dominant frequencies are extracted. The amplitude range (e.g. 0-5V), amplitude probability (e.g. the number of times the amplitude of 1V appears accounts for 20%), and frequency distribution (e.g. mainly concentrated in 200-500Hz) of these components are counted to form a "logistics sorting vibration statistical characteristic library". Noise signals that meet the above amplitude probability and frequency distribution are randomly generated and superimposed on the basic harmonic component obtained in step 221. At this time, the cuckoo algorithm is introduced to adjust the amplitude of the noise component. The scaling ratio of the overall amplitude of the noise is regarded as a "bird nest". The difference between the total energy of the superimposed signal and the total energy of the actual working condition is used as the evaluation standard. The initial scaling ratio is set to 1.0. When the difference exceeds 5%, the scaling ratio is adjusted. The 25% of the scaling ratios with the smallest differences are retained, and the remaining scaling ratios are randomly disturbed (e.g. a new ratio is generated within the range of 0.8-1.2). This simulates the selection of the bird nest by the cuckoo. When the difference between the total energy of the superimposed signal and the total energy of the actual working condition is less than 2%, the iteration is stopped. Finally, energy normalization is performed to calculate the ratio of the total energy of the superimposed signal to the total energy of the actual working condition. The amplitude of each time point in the signal is multiplied by the inverse of the ratio to ensure that the total energies of the two are completely consistent.

[0086] Determine the full working frequency range of the logistics sorting equipment (such as 5-1000Hz), divide it into several sub-frequency bands (such as 5-50Hz, 51-200Hz, 201-1000Hz), and count the energy proportion of each sub-frequency band for the random vibration excitation signal adjusted in step 222. The energy proportion of each sub-frequency band and the waveform change rate are regarded as "bird nest parameters", and "whether the energy proportion of each sub-frequency band covers the working frequency range of the equipment and is consistent with the actual working condition" is taken as the judgment standard. In the initial stage, the signal energy is distributed according to the energy proportion of each sub-frequency band in the actual working condition; when the algorithm runs, the un-covered frequency band (such as the actual 200-500Hz energy but the signal is missing) is adjusted specifically, the parameters of the covered sub-frequency band are retained, and the parameters of the missing frequency band are updated - the energy value is randomly generated in the missing frequency band (such as supplementing 5% vibration of the total energy in 200-500Hz), and the energy proportion of the adjacent frequency band is adjusted to keep the total energy unchanged. After 15 iterations, the generated signal has energy distribution in each sub-frequency band, and the waveform change with time is consistent with the fluctuation rule of the vibration intensity in the actual working condition, and finally a "wideband random vibration time-varying waveform covering the full working frequency range" is formed, which perfectly simulates the excitation environment of multi-frequency superposition.

[0087] The cuckoo algorithm can accurately lock the core frequency with the largest vibration energy contribution by multiple iterations of screening the dominant vibration frequency, avoiding the subjectivity of manual screening. At the same time, the updating mechanism of "inferior bird nest" can effectively exclude interference frequencies, ensuring that the identified dominant frequency, energy distribution and time change rule are highly consistent with the actual working condition. The parameter optimization (amplitude, initial phase) of the basic harmonic component can reduce the error between the synthesized harmonic signal and the actual vibration signal to a very low level through the "Levy flight" type fine-tuning mechanism. Combined with the energy proportion relationship, the basic harmonic generated not only retains the energy proportion characteristics of each frequency, but also improves the authenticity of the signal through algorithm optimization. The adjustment of random noise amplitude can accurately match the total energy level of the actual working condition, avoiding signal distortion caused by excessive or insufficient noise energy. Energy normalization processing combined with iterative optimization of the algorithm can ensure that the signal after superimposing noise is consistent with the actual vibration in terms of overall energy and local fluctuation, making the signal contain not only regular harmonics but also random components consistent with statistical characteristics, which is closer to the real complex vibration environment. The adjustment of the full working frequency range covers the missing frequency band to ensure that the generated wideband time-varying waveform has no frequency band omission, and the dynamic adjustment of the energy of each sub-frequency band can make the change rule of the signal with time completely synchronized with the fluctuation of the vibration in the actual working condition. The generated simulation signal can accurately reproduce the stress environment of the electromagnetic force balancing device in the actual sorting, providing simulation input for sensor calibration and improving the effectiveness and pertinence of calibration.

[0088] In a preferred embodiment of the present application, the step 3, based on the random vibration excitation signal, synchronously collects the electromagnetic force signal, the real-time acceleration signal of the measured object and the vibration frequency spectrum data of the sensor output in the calibration process of the electromagnetic force weighing sensor, and records the amplitude-frequency characteristic changes of the electromagnetic force under different vibration components, which can include:

[0089] Step 330, load the random vibration excitation signal to the electromagnetic force balancing device, use the acceleration sensor to collect the three-dimensional vibration data of the measured object in real time, and use the current detection unit to collect the current signal of the electromagnetic coil in real time;

[0090] Step 331, based on the three-dimensional vibration data, calculate the vibration frequency spectrum data of the measured object;

[0091] Step 332, based on the real-time current signal of the electromagnetic coil and the physical parameters of the electromagnetic force weighing sensor, calculate the electromagnetic force signal;

[0092] Step 333, perform time-frequency correlation analysis on the vibration frequency spectrum data and the electromagnetic force signal, and identify the amplitude variation characteristics of the electromagnetic force signal under different vibration frequency components, to obtain the correlation analysis result;

[0093] Step 334, based on the correlation analysis result, establish a mapping relationship table of vibration frequency and electromagnetic force amplitude response, and record the variation of the electromagnetic force amplitude under the vibration frequency components in the full frequency band.

[0094] In the embodiment of the present application, the constructed random vibration excitation signal is connected to the vibration driving module of the electromagnetic force balancing device, the output power of the module is adjusted to make the vibration intensity generated by the device consistent with the maximum vibration amplitude of the logistics sorting site (such as the amplitude range is controlled within 0.1-5 millimeters); a high-precision acceleration sensor is fixed on the top, side and bottom center of the measured object, the measuring axes of the three sensors are perpendicular to each other, corresponding to the horizontal front and rear (X-axis), horizontal left and right (Y-axis) and vertical up and down (Z-axis) directions respectively, the sampling frequency of the sensor is set to 1000 Hz, that is, 1000 groups of data are collected per second, to ensure that high-frequency vibration details can be captured.

[0095] Each sensor will convert the acceleration physical quantity generated by vibration into a voltage signal (such as 1 meter / second2change in acceleration corresponding to 0.1 volt change in voltage) and attach a time stamp accurate to milliseconds when collecting, continuously collect for 10 minutes, form a three-dimensional vibration data table containing X-axis, Y-axis, Z-axis real-time acceleration values and corresponding time, each row in the table represents three-direction acceleration information at a time point, at the same time, a shunt is connected in the power supply circuit of the electromagnetic coil, the resistance value of the shunt is accurately calibrated (such as 0.01 ohm), the real-time current is calculated by measuring the voltage difference across the shunt (the voltage difference is proportional to the current), the sampling frequency of the current detection unit is also set to 1000 Hz, which is strictly synchronized with the acceleration sensor (triggered by the same clock source), and the current value is immediately converted when each voltage difference value is collected (such as 0.001 volt voltage difference corresponding to 0.1 ampere current), forming a current signal sequence with a time stamp, ensuring that the current data at each time point can be accurately matched with the three-dimensional vibration data.

[0096] The three-dimensional vibration data is pre-processed, first the abnormal values in each direction acceleration signal that exceed the normal range are selected (such as the normal range of X-axis is-10 to 10 meter / second2, values exceeding this range are marked as abnormal), then the average value of the adjacent 5 data before and after the abnormal value is used to replace the abnormal value, to eliminate the influence of sensor burst interference, the processed X-axis, Y-axis, Z-axis data are respectively split into multiple continuous data blocks in time sequence, each data block contains 2048 data points (corresponding to 2.048 seconds); for each data block, calculate all the vibration frequency components contained therein, starting from 1 Hz, set a frequency point every 0.5 Hz, up to 2000 Hz, for each frequency point, determine its intensity by counting the acceleration amplitude of the vibration at that frequency, for example, calculate the acceleration amplitude (such as 2.5 meter / second2) caused by the vibration of the data block at 10 Hz.

[0097] All frequency points and their corresponding acceleration amplitudes of each data block are arranged into a frequency spectrum segment for that period, and all frequency spectrum segments are arranged in time sequence to form complete vibration spectrum data, which not only reflects the vibration intensity of each frequency, but also shows the intensity change of the same frequency at different time periods (such as 10 Hz vibration intensity is 2.5 meter / second2at 3 minutes, 1.8 meter / second2at 5 minutes).

[0098] Detailed physical parameters of the electromagnetic force weighing sensor are collected, the number of turns of the electromagnetic coil is 500 turns, the effective length of the coil in the magnetic field is 0.05 meters, the magnetic induction intensity of the working magnetic field is 0.2 Tesla (these parameters are obtained through the sensor factory calibration report), the current value at each time point (such as the current at a certain time point is 2 amperes) is extracted from the current signal sequence of step 330, the electromagnetic force at the time point is calculated according to the above physical parameters, and the electromagnetic force, current, coil parameters are combined to obtain the electromagnetic force corresponding to the current (such as 2 amperes corresponding to the electromagnetic force of 10 newtons).

[0099] The above calculation is performed for each time point, and the results are arranged in chronological order to form an electromagnetic force signal, each data point in the signal is attached with a time stamp, and the dynamic change of the electromagnetic force with time can be clearly displayed (such as gradually increasing from 8 newtons at 1 second to 12 newtons at 2 seconds).

[0100] The vibration spectrum data and the electromagnetic force signal are aligned according to the time stamp to ensure that the vibration spectrum and the electromagnetic force data at the same time point match each other, for example, at 5.000 seconds, the acceleration amplitude of 50 Hz in the vibration spectrum is 3 meters / second2, and the electromagnetic force at this time is 15 newtons, for each frequency point (from 1 Hz to 2000 Hz, interval 0.5 Hz), the acceleration amplitude of the frequency at all time points and the electromagnetic force amplitude at the corresponding time point are extracted to form the "vibration-electromagnetic force" correlation data set of the frequency, the data set is analyzed, the average change amount of the electromagnetic force amplitude when the acceleration amplitude of the frequency changes by 1 meter / second2 is calculated (such as when the acceleration of 50 Hz increases by 1 meter / second2, the electromagnetic force increases by 0.8 newtons on average), and whether the change is stable (such as the fluctuation range of the change amount is 0.7-0.9 newtons) is recorded; the analysis results of all frequency points are summarized to form the correlation analysis results, and the influence characteristics of different frequency vibrations on the electromagnetic force amplitude are determined (such as low-frequency vibration has more stable influence on electromagnetic force, and high-frequency vibration has greater fluctuation).

[0101] According to the correlation analysis result, a record is established for each frequency point (1 Hz to 2000 Hz, 0.5 Hz interval), including the frequency value, the average change amount of the electromagnetic force corresponding to the change of 1 m / s2 of the vibration acceleration at the frequency, the maximum deviation value of the change amount, and the time proportion (such as 100 Hz vibration accounts for 15% in the total collection time) of the vibration at the frequency, the records are arranged in order of frequency from low to high to form a mapping relationship table of vibration frequency and electromagnetic force amplitude response, the table can not only query the influence degree of any frequency on the electromagnetic force, but also reflect the influence difference of the same frequency under different vibration intensities (such as when the 100 Hz vibration acceleration is 2 m / s2, the electromagnetic force changes by 1.5 N; when the acceleration is 5 m / s2, the electromagnetic force changes by 3.8 N), through the table, the specific influence of all vibration frequency components in the full frequency band on the electromagnetic force amplitude can be comprehensively mastered, including the size, stability and occurrence frequency of the influence.

[0102] By synchronously collecting three-dimensional vibration data and current signals, and using high-precision sensors and strict synchronization mechanism, the accurate correspondence of vibration and electromagnetic force related data in time and space is ensured, and the analysis deviation caused by different data is avoided; the pre-processing of vibration data and detailed spectrum analysis can accurately extract the intensity and change rule of each frequency vibration, clearly present the vibration characteristics of the measured object in different directions and different frequencies, and provide detailed basis for identifying key influence frequencies; the electromagnetic force signal is calculated based on the current signal and the physical parameters of the sensor, the abstract current change is converted into intuitive electromagnetic force value, and the calculation process combines the inherent properties of the sensor, so that the accuracy and reliability of the electromagnetic force signal are ensured, a direct quantitative index for studying the influence of vibration on electromagnetic force is provided, the time-frequency correlation analysis can analyze the relationship between each frequency vibration and electromagnetic force one by one, the specific influence degree and rule of different frequencies are determined, the ambiguity of overall analysis is avoided, the action mechanism of vibration on electromagnetic force is clear, and a clear foundation is laid for establishing the mapping relationship; the mapping relationship table comprehensively records the influence of full-band vibration on electromagnetic force, including the size, stability and occurrence frequency of the influence, so that the staff can quickly query the influence of any frequency, and the comprehensive compensation mechanism provides detailed reference.

[0103] In a preferred embodiment of the present application, the step 4 of extracting the nonlinear distortion parameters corresponding to each vibration frequency band according to the amplitude-frequency characteristic change of the electromagnetic force to form the comprehensive compensation mechanism can include:

[0104] Step 440, based on the vibration frequency-electromagnetic force amplitude response mapping table, analyzing the gain change condition of the full-band electromagnetic force, and identifying the abnormal vibration frequency band deviating from the linear theoretical value of the gain through a preset gain deviation threshold value;

[0105] Step 441, for the abnormal vibration frequency band, the dynamic deviation amount of the electromagnetic force theoretical prediction value and the sensor measured value in the frequency band is calculated, and a frequency-indexed nonlinear distortion parameter set is generated;

[0106] Step 442, the frequency-indexed nonlinear distortion parameter set is divided into a segmented frequency domain compensation rule set according to the abnormal frequency band boundary frequency division rule, and a dynamic force real-time compensation mechanism is formed.

[0107] In the embodiment of the application, for "analyzing the gain variation of the electromagnetic force in the whole frequency band based on the vibration frequency-electromagnetic force amplitude response mapping table", the basic data of vibration frequency and corresponding electromagnetic force amplitude are collected first, which come from the previous experiment or equipment operation record, covering the whole vibration frequency range from the lowest to the highest, for example, if the frequency range is set to 20Hz to 2000Hz, the corresponding electromagnetic force amplitude of every 1Hz in this range needs to be collected to form the vibration frequency-electromagnetic force amplitude response mapping table, when analyzing, the electromagnetic force amplitude of the first frequency point (20Hz) is extracted from the table, and the electromagnetic force amplitude of the next frequency point (21Hz) is extracted, the variation of the electromagnetic force amplitude between the two frequency points is obtained by subtracting the former from the latter, in this way, the variation between all adjacent frequency points is calculated in turn, by comparing the size of these variations, it is determined whether the gain is rising, falling or remaining stable, for example, the amplitude corresponding to 20Hz is 5N, the amplitude corresponding to 21Hz is 5.2N, the variation is 0.2N; the variation from 21Hz to 22Hz is 0.3N, which indicates that the gain in this frequency range is rising and the rising amplitude is increasing.

[0108] For "obtaining the abnormal vibration frequency band deviating from the linear theoretical value of the gain through the preset gain deviation threshold", first, the preset gain deviation threshold is determined, for example, the difference between the actual value and the theoretical value cannot exceed ±0.3N, then, according to the linear theory, each frequency point has a corresponding electromagnetic force theoretical gain value, which is a fixed value derived through ideal linear mechanism, then, the actual electromagnetic force gain value of each frequency point is taken from the mapping table, and the theoretical value of the frequency point is subtracted, if the operation result (actual value minus theoretical value) of a certain frequency point is greater than 0.3N or less than -0.3N, the frequency point is marked as abnormal, when multiple abnormal frequency points appear continuously, the frequency of the first abnormal point is taken as the starting boundary, and the frequency of the last abnormal point is taken as the ending boundary, the range between the two boundaries is the abnormal vibration frequency band, for example, from 100Hz to 110Hz, the difference between the actual value and the theoretical value of each frequency point exceeds 0.3N, so 100Hz to 110Hz is the abnormal vibration frequency band.

[0109] For "calculating the dynamic deviation amount of the electromagnetic force theoretical prediction value and the sensor measured value in the frequency band", for the determined abnormal vibration frequency band, such as 100Hz to 110Hz, first determine the electromagnetic force theoretical prediction value of each frequency point. This value is a fixed value calculated according to the linear theory formula combined with the parameters of the frequency point. At the same time, the electromagnetic force of each frequency point in the frequency band is measured by the sensor for multiple times. For example, at 100Hz, each is measured once at 1 minute, 2 minutes and 3 minutes after the device starts, and three measured values are obtained. Then, subtract the measured value of each time point from the theoretical prediction value of each time point to obtain the deviation amount of each time point. For example, the theoretical prediction value of 100Hz is 10N, the measured value at 1 minute is 9.5N, the deviation amount is 0.5N; the measured value at 2 minutes is 9.4N, the deviation amount is 0.6N; the measured value at 3 minutes is 9.6N, the deviation amount is 0.4N. Arrange these deviation amounts in time sequence to see the dynamic change of the deviation amount of the frequency point, that is, the dynamic deviation amount.

[0110] For "generating a set of nonlinear distortion parameters indexed by frequency", take each specific frequency in the abnormal vibration frequency band (such as 100Hz, 101Hz…110Hz) as the index keyword, and collect all the dynamic deviation amounts corresponding to each frequency point (including deviation values at different time points) and associate them with the frequency index. For example, the dynamic deviation amount corresponding to 100Hz is 0.5N, 0.6N, 0.4N; the dynamic deviation amount corresponding to 101Hz is 0.55N, 0.65N, 0.45N, etc. Integrate these data classified by frequency index to form a set of nonlinear distortion parameters.

[0111] For "establishing a segmented frequency domain compensation rule set according to the abnormal frequency band boundary frequency division rule", first, the abnormal frequency band boundary frequency division rule is determined, for example, when the average value of the dynamic deviation of the two adjacent frequency points is more than 0.2N, the two frequency points are taken as the segment boundary, then the frequency index type nonlinear distortion parameter set is checked, the average value of the dynamic deviation of each frequency point is calculated (for example, the average value of 100Hz is (0.5+0.6+0.4)÷3=0.5N, the average value of 101Hz is (0.55+0.65+0.45)÷3=0.55N, the difference is 0.05N, which does not meet the boundary condition; while the average value of 105Hz is 0.8N, and the average value of 106Hz is 1.1N, the difference is 0.3N, which exceeds 0.2N, so 105Hz and 106Hz are taken as the boundary), after the abnormal frequency band is divided into multiple segments according to this rule, the dynamic deviation range and change trend of all frequency points in each segment are counted, and the compensation rule is formulated, for example, for the 100Hz to 105Hz segment, the dynamic deviation is between 0.4-0.8N and gradually increases, the rule of "in this segment, the compensation amount increases by 0.05N for every 1Hz increase in frequency" is formulated, and the rule set of the segmented frequency domain compensation rule set is formed after the rules of all segments are summarized.

[0112] For "forming a dynamic force real-time compensation mechanism", during the operation of the device, the current vibration frequency is monitored in real time, when the frequency falls into a certain segment (such as 100Hz to 105Hz), the compensation rule corresponding to the segment is called, and the compensation value of the electromagnetic force is calculated according to the specific position of the current frequency in the segment according to the rule, for example, the current frequency is 103Hz, which is in the 100Hz to 105Hz segment, the initial compensation amount of the segment is 0.4N, and the compensation amount increases by 0.05N for every 1Hz increase, so the compensation amount corresponding to 103Hz is 0.4+(103-100)×0.05=0.55N, then the compensation amount is superimposed on the original electromagnetic force in real time, and the compensation amount is continuously updated with the change of the frequency, forming a dynamic force real-time compensation mechanism.

[0113] By analyzing the electromagnetic force gain change of the full frequency band point by point and combining the preset threshold to identify the abnormal frequency band, the problem area can be accurately locked, the invalid processing of the normal frequency band is avoided, and the unnecessary calculation and operation cost are reduced; the dynamic deviation amount is calculated for the abnormal frequency band, and a set of nonlinear distortion parameters is generated, which records the distortion of each frequency point in detail, provides specific data basis for compensation, ensures that the compensation measures can fit the actual distortion state, and improves the accuracy of compensation; the segmented compensation rule set is established and the dynamic real-time compensation mechanism is formed, which can flexibly adjust the compensation strategy according to the characteristics of different frequency bands, respond to the distortion change caused by the frequency change in real time, reduce the influence of nonlinear distortion on the equipment, and improve the stability and accuracy of the output force of the equipment.

[0114] In a preferred embodiment of the present application, the above step 5, three detection points with fixed spatial positions are arranged on the logistics sorting line, and the vibration characteristics of each detection point are obtained in combination with the comprehensive compensation mechanism, and based on the spatial distribution relationship of the three detection points, a dynamic compensation coefficient is generated, which can include:

[0115] Step 550, three vibration sensors are linearly arranged as detection points at a preset fixed interval along the running path of the conveying belt in the bearing direction of the weighing station, the first detection point is located at the starting section of the conveying belt, the second detection point is located at the middle section of the conveying belt, and the third detection point is located at the end section of the conveying belt, and the frequency spectrum distribution data of the acceleration signal of the conveying belt at each detection point is collected in real time;

[0116] Step 551, combine the frequency spectrum distribution data of the three detection points with the corresponding spatial coordinate values, and obtain the rigid vibration component data and flexible deformation component data of the conveying belt according to the predefined vibration component separation rule;

[0117] Step 552, based on the rigid vibration component data and the flexible deformation component data, calculate the dynamic influence ratio at the weighing position, and generate a set of adaptive compensation coefficients bound to the spatial position.

[0118] In the embodiment of the present application, when determining the preset fixed interval, first measure the actual total length of the conveying belt from the starting end to the end, for example, the total length is measured to be 15 meters, then combine the vibration propagation characteristics of the conveying belt during operation, assume that setting a detection point every 5 meters can better cover the vibration at different positions, then set the preset fixed interval to 5 meters, and then determine the specific positions of the three detection points in turn, the first detection point is at the starting section position 0 meters away from the starting end of the conveying belt, the second detection point is at the middle section position 5 meters away from the starting end, and the third detection point is at the end section position 10 meters away from the starting end (the remaining 5 meters of the end section is used as a buffer to ensure that the detection point can reflect the end vibration).

[0119] When collecting acceleration signals in real time, each vibration sensor will collect 2000 acceleration values per second. For example, the values collected by the first detection point in a second are 0.2 m / s 2 , 0.3 m / s 2 , 0.25 m / s 2 …… These values are arranged in the order of collection time to form a continuous acceleration signal sequence. When obtaining frequency spectrum distribution data, the acceleration signal sequences of each detection point for 10 seconds are integrated to obtain a signal set containing 20000 data, and then the intensities corresponding to different vibration frequencies in these signals are observed. For example, it is found that the average intensity of the acceleration signal at a frequency of 10 Hz is 0.15 m / s 2 , and the average intensity at a frequency of 20 Hz is 0.1 m / s 2 …… In this way, the acceleration signal intensity corresponding to each frequency is sorted out to form the frequency spectrum distribution data of each detection point.

[0120] The frequency spectrum distribution data of the three detection points and the corresponding spatial coordinate values are combined to obtain the rigid vibration component data and the flexible deformation component data of the conveyor belt according to the predefined vibration component separation rule. When combining the frequency spectrum distribution data and the spatial coordinate values, the spatial coordinates of the three detection points are first determined. Assuming that the starting end of the conveyor belt is the origin and the bearing direction is the x-axis, the coordinates of the first detection point are (0, 0, 0), the coordinates of the second detection point are (5, 0, 0), and the coordinates of the third detection point are (10, 0, 0). Then, the frequency spectrum distribution data of each detection point is mapped to the respective coordinates. For example, the acceleration intensity of the first detection point (0, 0, 0) at 10 Hz is 0.15 m / s 2 , and the acceleration intensity at 20 Hz is 0.1 m / s 2 …… The data are bound to the coordinate (0, 0, 0). The frequency spectrum data of the second detection point (5, 0, 0) and the third detection point (10, 0, 0) are also bound to their respective coordinates in the same way.

[0121] According to the predefined vibration component separation rule, when the difference in acceleration intensity of the three detection points at the same frequency does not exceed 0.03 m / s 2 , it is determined to be a rigid vibration component; when the difference exceeds 0.03 m / s 2 , it is determined to be a flexible deformation component. For example, at a frequency of 30 Hz, the intensity of the first detection point is 0.2 m / s 2 , the intensity of the second detection point is 0.22 m / s 2 , and the intensity of the third detection point is 0.21 m / s 2 . The maximum difference between the three values is 0.02 m / s 2 , which does not exceed 0.03 m / s 2Therefore, the vibration component at this frequency is classified as rigid vibration component data; while at 40Hz, the first detection point intensity is 0.1m / s 2 , the second detection point is 0.15m / s 2 , the third detection point is 0.2m / s 2 , the maximum difference is 0.1m / s 2 , and more than 0.03m / s 2 , it is classified as flexible deformation component data. In this way, two component data are separated from the spectrum data of all frequencies.

[0122] When calculating the dynamic influence ratio at the weighing position, the coordinates of the weighing position are determined first. It is assumed that the weighing position is at a distance of 7 meters from the starting end, and the coordinates are (7, 0, 0). Then the rigid vibration component data and the flexible deformation component data corresponding to this position are extracted, and the total acceleration intensity of the rigid vibration component at each frequency is calculated, for example, in the range of 10Hz to 50Hz, the total intensity of the rigid vibration at each frequency is 1.2m / s 2 ; the total intensity of the flexible deformation component at each frequency is 0.8m / s 2 . Then the dynamic influence ratio is calculated, the rigid vibration component influence ratio is 1.2 ÷ (1.2 + 0.8) = 0.6, and the flexible deformation component influence ratio is 0.8 ÷ (1.2 + 0.8) = 0.4.

[0123] When generating the adaptive compensation coefficient set, the compensation coefficient is set according to the dynamic influence ratio. For the rigid vibration component, according to its influence ratio of 0.6, when this component is detected, the adjustment amplitude of the weighing result is set to 6% of the actual weighing value as compensation; the flexible deformation component influence ratio is 0.4, and the adjustment amplitude is set to 4% of the actual weighing value as compensation. Bind these two compensation coefficients with the weighing position (7, 0, 0) to form the compensation coefficient of this position. In the same way, for other spatial positions on the conveyor belt that need to be concerned, such as (2, 0, 0), (9, 0, 0), etc., the respective rigid and flexible component influence ratios are calculated, the corresponding compensation coefficients are set, and finally all the position compensation coefficients are integrated to form the adaptive compensation coefficient set bound with the spatial position.

[0124] By accurately setting the detection point spacing and collecting detailed spectrum data, the vibration intensity of the conveyor belt at different positions under various frequencies can be comprehensively captured, avoiding one-sided vibration analysis due to insufficient data. At the same time, real-time data collection ensures the timeliness of the data, which can timely reflect the real-time changes of the conveyor belt vibration. By combining the spectrum data with the spatial coordinates and separating the vibration components according to clear rules, the influence of rigid and flexible vibrations is clearly distinguished, allowing staff to analyze the characteristics of different types of vibrations and avoid ambiguous judgments of the vibration source. According to the dynamic influence ratio, a set of compensation coefficients bound to the spatial position is generated, allowing compensation at each position to be based on the actual vibration of the position, improving the accuracy of compensation. For example, different positions have different rigid and flexible vibration influences, and the compensation coefficient adjusts accordingly, effectively reducing the interference of vibration on the weighing result and improving the weighing accuracy. The three steps form a complete dynamic compensation process, from data collection to compensation implementation, which focuses on specific positions and vibration types, making the weighing of the logistics sorting line less affected by vibration and improving the sorting efficiency and weighing accuracy. At the same time, the adaptive compensation method can adapt to the vibration changes under different running states of the conveyor belt, enhancing stability and applicability.

[0125] In a preferred embodiment of the present application, the above step 6, according to the real-time vibration characteristics and dynamic compensation coefficients in logistics sorting, dynamically adjusts the matching relationship of each frequency band parameter in the comprehensive compensation mechanism, which can include:

[0126] Step 660, input the real-time collected vibration spectrum data into the segmented frequency domain compensation unit, and call the dynamic compensation coefficient set to reconstruct the compensation weight parameters of each frequency band;

[0127] Step 661, based on the reconstructed compensation weight parameters, calculate the proportion of high-frequency vibration energy in the total spectrum, and generate a high-frequency compensation intensity enhancement coefficient according to the proportion value;

[0128] Step 662, extract the low-frequency amplitude offset in the compensation weight parameter, and combine the high-frequency compensation intensity enhancement coefficient to dynamically correct the frequency band parameters in the compensation unit.

[0129] In the embodiment of the present application, the vibration frequency spectrum data collected in real time covers vibration information from 5 Hz to 2000 Hz, each frequency point (such as 5 Hz, 6 Hz,..., 2000 Hz) has a corresponding vibration amplitude, for example, 5 Hz corresponds to 0.02 mm, 6 Hz corresponds to 0.023 mm, and so on; after inputting these data into the segmented frequency domain compensation unit, the compensation unit divides the frequency bands according to the preset rules, for example, 5-200 Hz is the low frequency band, 201-800 Hz is the medium frequency band, and 801-2000 Hz is the high frequency band; in the called dynamic compensation coefficient set, each frequency band has a basic compensation coefficient, such as the low frequency band basic coefficient 0.25, the medium frequency band 0.45, and the high frequency band 0.3, at the same time, the coefficient set also contains the historical vibration fluctuation range of each frequency band, for example, the normal fluctuation of the low frequency band is between 15-25 mm, when calculating the average amplitude of each frequency band, the low frequency band contains 196 frequency points of 5-200 Hz, the amplitudes of each point are added, for example, 5 Hz is 0.02 mm, 6 Hz is 0.023 mm,..., 200 Hz is 0.05 mm, the total is 12.5 mm, and the average amplitude is 12.5 mm divided by 196 ≈ 0.0638 mm, the medium frequency band average amplitude is calculated by the same method 0.085 mm, and the high frequency band average amplitude is calculated by the same method 0.042 mm.

[0130] The preliminary frequency band compensation value is calculated, the low frequency band is 0.0638 mm multiplied by 0.25 ≈ 0.01595; the medium frequency band is 0.085 mm multiplied by 0.45 ≈ 0.03825; and the high frequency band is 0.042 mm multiplied by 0.3 ≈ 0.0126. When normalized, the total is 0.01595+0.03825+0.0126 ≈ 0.0668, the low frequency band weight is 0.01595 ÷ 0.0668 ≈ 0.239, the medium frequency band is 0.03825 ÷ 0.0668 ≈ 0.573, and the high frequency band is 0.0126 ÷ 0.0668 ≈ 0.189.

[0131] The high frequency vibration energy is calculated, the amplitude sum of all frequency points in the high frequency band (801-2000 Hz) is first counted, for example, the sum of the amplitudes of each point is 8.5 mm, which is multiplied by the high frequency compensation weight 0.189 to obtain 8.5 × 0.189 ≈ 1.6065; the total frequency spectrum energy is calculated, the low frequency band amplitude sum 6.2 mm is multiplied by the weight 0.239 to obtain 1.4818; the medium frequency band amplitude sum 9.3 mm is multiplied by the weight 0.573 to obtain 5.3289; and the high frequency energy 1.6065 is added to obtain the total energy 4818+5.3289+1.6065 ≈ 8.4172; the high frequency proportion is 1.6065 ÷ 8.4172 ≈ 0.1908 (19.08%), when the enhancement coefficient is generated, the basic coefficient is set to 1.0, the part exceeding 10% is enhanced by 0.05 for each 1%, and the enhancement coefficient is 1.0+9.08 × 0.05 ≈ 1.454.

[0132] Extract the low-frequency amplitude offset, the preset standard low-frequency average amplitude is 0.05mm, the current actual average amplitude is 0.0638mm, the offset is 0.0638-0.05=0.0138mm, correct the high-frequency segment parameter, the original high-frequency weight is 0.189 multiplied by the enhancement coefficient 1.454≈0.2748; correct the low-frequency segment parameter, because the offset is positive, reduce the weight 0.02 every 0.01mm of offset, 0.0138mm corresponds to reduce 0.02×1.38≈0.0276, the corrected low-frequency weight is 0.239-0.0276≈0.2114; correct the mid-frequency segment parameter, the total weight remains 1, and the new weight of the mid-frequency segment is 1-0.2114-0.2748≈0.5138.

[0133] The calculation process is detailed to each frequency point, so that the compensation parameter can accurately reflect the subtle changes of real-time vibration, ensure that the compensation mechanism is highly matched with the actual vibration state, and reduce the error caused by the mismatch of the sorting equipment due to vibration; the accurate calculation of the proportion of high-frequency vibration energy and the targeted generation of the enhancement coefficient can effectively reduce the interference of high-frequency vibration on the sorting accuracy, and protect the key components of the equipment from excessive damage caused by high-frequency vibration; the extraction of the low-frequency amplitude offset and the dynamic correction of the parameters of each frequency segment realize the intelligent balance of the compensation intensity of different frequency segments, avoid overcompensation or insufficient compensation of a certain frequency segment, and improve the stability and adaptability of the entire compensation system; the specific calculation based on real-time data in the whole process makes the adjustment of the compensation parameter more scientific and traceable, and improves the logistics sorting efficiency.

[0134] In a preferred embodiment of the present application, the above step 7, based on the adjusted comprehensive compensation mechanism, adjusts the current of the electromagnetic coil to achieve dynamic balance between the electromagnetic force and the measured mass, and obtains the weighing accuracy under different working conditions, which can include:

[0135] Step 770, input the updated comprehensive compensation mechanism parameter set into the electromagnetic force control unit, combine the real-time collected vibration spectrum data to analyze the vibration interference force component, and generate a current compensation signal group distributed discretely according to the frequency band;

[0136] Step 771, extract the compensation amount of each frequency band in the current compensation signal group, and perform amplitude-phase synchronous superposition according to the frequency band with the reference driving current of the electromagnetic coil to generate an anti-vibration synthesized driving current instruction set;

[0137] Step 772, control the current of the electromagnetic coil through the anti-vibration synthesized driving current instruction set, and collect the displacement state sequence of the weighing unit by using the displacement sensing device;

[0138] Step 773, perform dynamic stationarity detection on the displacement state sequence, and when the displacement fluctuation amplitude is continuously within the zeroing tolerance range for a preset period, the mass measurement value after vibration isolation is obtained.

[0139] In the embodiment of the present application, for "inputting the updated comprehensive compensation mechanism parameter set into the electromagnetic force control unit", the parameters in the updated comprehensive compensation mechanism parameter set are collected first, such as compensation coefficients of different vibration frequency bands, response speed parameters, etc., and then these parameters are input into the electromagnetic force control unit one by one, and the control unit will identify and store these parameters.

[0140] For "analyzing the vibration interference force component in combination with the real-time collected vibration frequency spectrum data", the vibration frequency spectrum data are collected in real time, which contain vibration amplitude information at different frequencies, the electromagnetic force control unit calls the parameters of the corresponding frequency band in the stored comprehensive compensation mechanism parameter set, multiplies the vibration amplitude at each frequency by the compensation coefficient of the frequency band, and adjusts the product result in time according to the response speed parameter, so as to obtain the size and change trend of the vibration interference force component of each frequency band.

[0141] For "generating the current compensation signal group distributed discretely by frequency band", according to the analyzed vibration interference force component of each frequency band, the electromagnetic force control unit converts the size of each component into the amplitude of the corresponding current compensation signal according to the preset conversion rule, at the same time, determines the change rate and duration of the current compensation signal according to the change trend of the vibration interference force component, and then generates the current compensation signal corresponding to each frequency band, and these signals combined together form the current compensation signal group distributed discretely by frequency band.

[0142] For "extracting the compensation amount of each frequency band in the current compensation signal group", the amplitude of the current compensation signal corresponding to each frequency band is extracted from the generated current compensation signal group, and this amplitude is the compensation amount of the frequency band, for example, the amplitude of the current compensation signal of a certain frequency band is stable at 5A within a period of time, and the compensation amount of the frequency band is 5A.

[0143] For "amplitude-phase synchronous superposition of each frequency band with the electromagnetic coil reference driving current", first, the amplitudes and phases of the electromagnetic coil reference driving current in each frequency band are determined, then the amplitude of the current compensation amount of each frequency band is added to the amplitude of the reference driving current in the same frequency band to obtain the superimposed amplitude, and in terms of phase, the phase of the current compensation amount is ensured to be consistent with the phase of the reference driving current in the same frequency band to achieve phase synchronization, and then the amplitude-phase synchronous superposition of each frequency band is completed.

[0144] For "generating the anti-vibration synthesized driving current instruction set", the current information of each frequency band after amplitude-phase synchronous superposition is integrated and arranged in the order of frequency band to form a complete set of anti-vibration synthesized driving current instructions, each instruction corresponding to the current output requirement of a frequency band.

[0145] For "controlling the electromagnetic coil current supply through the anti-vibration synthesis driving current instruction set", the anti-vibration synthesis driving current instruction set contains the current output value of each frequency band, and the control system will send the current value to the current supply device according to the order in the instruction set, and the current supply device will adjust its output after receiving the current value of each frequency band to ensure that the current provided for the electromagnetic coil reaches the required size in the frequency band, and sequentially complete the current supply control of all frequency bands; for "simultaneously collecting the displacement state sequence of the weighing unit by using the displacement sensing device", the displacement sensing device will collect the displacement of the load-bearing unit at fixed time intervals, such as once every 0.1 seconds, and each time the device will measure the displacement distance of the load-bearing unit relative to the initial position, and record these distances in sequence according to the collection time to form a displacement state sequence.

[0146] For "dynamic stationarity detection of the displacement state sequence", a displacement fluctuation detection period is set, such as 0.5 seconds as a period, and in each period, all displacement values in the period are extracted from the displacement state sequence, the maximum and minimum values are found, and the difference between the two is calculated, which is the displacement fluctuation amplitude in the period, then the fluctuation amplitudes of adjacent periods are compared to observe the changes and determine whether the displacement state is in a stationary state; for "when the displacement fluctuation amplitude is within the zeroing tolerance range for a continuous preset period, the mass measurement value after vibration isolation is obtained", first set the zeroing tolerance range, such as ±0.01mm, and the preset period, such as 3 periods, continuously monitor the displacement fluctuation amplitude of each period, and when it is found that the fluctuation amplitude of the last 3 periods is within the range of ±0.01mm, the mass value displayed by the current weighing unit at this time is recorded, which is the mass measurement value after vibration isolation.

[0147] By analyzing the vibration interference force components in detail and generating a targeted current compensation signal group, different frequency bands of vibration interference can be accurately identified, the compensation is more targeted, and the accuracy of the early preparation against vibration interference is improved; the amplitude-phase synchronous superposition of the compensation amount of each frequency band and the reference driving current ensures the accuracy and coordination of current superposition, avoids new interference caused by phase or amplitude mismatch, and the generated anti-vibration synthesis driving current instruction set can more effectively resist vibration influence, improving the anti-vibration ability of current driving; collecting the displacement state sequence while controlling the electromagnetic coil current supply realizes real-time monitoring of the entire weighing process, and can timely understand the displacement change of the weighing unit, facilitating timely problem discovery and adjustment; the dynamic stationarity detection determines the mass measurement value, ensuring that the measurement result is obtained only when the displacement is stable, reducing the measurement error caused by displacement fluctuation, improving the accuracy and reliability of mass measurement, and making the obtained mass measurement value closer to the true value.

[0148] AsFigure 2 As shown, the embodiment of the application also provides an electromagnetic force weighing dynamic calibration system, comprising:

[0149] The signal acquisition module is configured to acquire the composite vibration signal and form the working condition vibration characteristic data in the logistics sorting under the conditions of high-speed movement, frequent start-stop and random vibration of the package.

[0150] The signal construction module is configured to construct a random vibration excitation signal covering the full frequency band based on the working condition vibration characteristic data, so as to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition.

[0151] The synchronous acquisition module is configured to synchronously acquire the electromagnetic force signal output by the sensor, the real-time acceleration signal of the measured object and the vibration frequency spectrum data in the calibration process of the electromagnetic force weighing sensor based on the random vibration excitation signal, and record the amplitude-frequency characteristic variation of the electromagnetic force under different vibration components.

[0152] The parameter extraction module is configured to extract the nonlinear distortion parameters corresponding to each vibration frequency band according to the amplitude-frequency characteristic variation of the electromagnetic force, and form a comprehensive compensation mechanism.

[0153] The coefficient generation module is configured to set three detection points fixed in space on the logistics sorting line, acquire the vibration characteristics of each detection point in combination with the comprehensive compensation mechanism, generate a dynamic compensation coefficient based on the spatial distribution relationship of the three detection points.

[0154] The dynamic adjustment module is configured to dynamically adjust the matching relationship of the parameters in each frequency band in the comprehensive compensation mechanism according to the real-time vibration characteristics in the logistics sorting and the dynamic compensation coefficient.

[0155] The balance adjustment module is configured to adjust the current of the electromagnetic coil based on the adjusted comprehensive compensation mechanism, so as to realize the dynamic balance between the electromagnetic force and the measured mass, and obtain the weighing precision under different working conditions.

[0156] It should be noted that the system corresponds to the above method, and all the implementation manners in the above method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0157] The embodiment of the application also provides a computing device, comprising a processor and a memory storing a computer program, wherein the computer program is run by the processor to execute the method as described above. All the implementation manners in the above method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0158] The embodiment of the present application also provides a computer readable storage medium, which stores instructions, and when the instructions are run on a computer, the computer executes the method as described above. All implementation manners in the above method embodiment are suitable for this embodiment and can achieve the same technical effects.

[0159] The above is the preferred embodiment of the present application. It should be pointed out that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A dynamic calibration method of electromagnetic force weighing, characterized in that, The method comprises: Step 1, in the logistics sorting, the composite vibration signal is collected under the conditions of high-speed movement, frequent start-stop and random vibration of the package, and working condition vibration characteristic data is formed; Step 2, based on the working condition vibration characteristic data, a random vibration excitation signal covering the full frequency band is constructed to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition; Step 3, based on the random vibration excitation signal, the electromagnetic force signal output by the sensor, the real-time acceleration signal of the measured object and the vibration frequency spectrum data are synchronously collected in the calibration process of the electromagnetic force weighing sensor during the calibration process of the electromagnetic force weighing sensor, and the amplitude-frequency characteristic change of the electromagnetic force under different vibration components is recorded; Step 4, according to the amplitude-frequency characteristic change of the electromagnetic force, the nonlinear distortion parameters corresponding to each vibration frequency band are extracted to form a comprehensive compensation mechanism; Step 5, three detection points with fixed spatial positions are set on the logistics sorting line, and the vibration characteristics of each detection point are obtained in combination with the comprehensive compensation mechanism, and the dynamic compensation coefficient is generated based on the spatial distribution relationship of the three detection points; Step 6, according to the real-time vibration characteristics and the dynamic compensation coefficient in the logistics sorting, the matching relationship of the parameters in the comprehensive compensation mechanism is dynamically adjusted; Step 7, based on the adjusted comprehensive compensation mechanism, the current of the electromagnetic coil is adjusted to realize the dynamic balance of the electromagnetic force and the measured mass, and the weighing precision under different working conditions is obtained.

2. The electromagnetic force weighing dynamic calibration method of claim 1, wherein, Based on the working condition vibration characteristic data, a random vibration excitation signal covering the full frequency band is constructed to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition, which comprises: The time-frequency domain joint analysis is performed on the collected working condition vibration characteristic data, the dominant vibration frequency component, the energy distribution characteristics of each frequency component and the change rule of vibration energy with time under the conditions of high-speed movement, frequent start-stop and random vibration of the package in the logistics sorting process are identified; Based on the dominant vibration frequency component, the energy proportion relationship of each frequency component in the actual working condition is calculated based on the energy distribution characteristics of each frequency component, and a basic harmonic component is generated; In the basic harmonic component, the random noise component consistent with the statistical characteristics of the logistics sorting vibration is superimposed, the amplitude of the random noise component is adjusted, and the total energy level is matched with the energy of the actual working condition vibration characteristic data through energy normalization processing; Based on the adjusted random vibration excitation signal, a wideband random vibration time-varying waveform covering the full working frequency band of the logistics sorting equipment is constructed to generate a signal simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition.

3. The electromagnetic force weighing dynamic calibration method of claim 2, wherein, Based on the random vibration excitation signal, the electromagnetic force signal output by the sensor, the real-time acceleration signal of the measured object and the vibration frequency spectrum data are synchronously collected in the calibration process of the electromagnetic force weighing sensor during the calibration process of the electromagnetic force weighing sensor, and the amplitude-frequency characteristic change of the electromagnetic force under different vibration components is recorded, which comprises: The random vibration excitation signal is loaded to the electromagnetic force balancing device, the three-dimensional vibration data of the measured object are collected in real time by using the acceleration sensor, and the current signal of the electromagnetic coil is collected in real time by using the current detection unit; Based on the three-dimensional vibration data, the vibration frequency spectrum data of the measured object are calculated; The electromagnetic force signal is calculated based on the real-time current signal of the electromagnetic coil and the physical parameters of the electromagnetic force weighing sensor; The vibration frequency spectrum data and the electromagnetic force signal are subjected to time-frequency correlation analysis, and the amplitude variation characteristics of the electromagnetic force signal under different vibration frequency components are identified to obtain a correlation analysis result; Based on the correlation analysis result, a mapping relationship table of vibration frequency and electromagnetic force amplitude response is established to record the variation of the electromagnetic force amplitude caused by the vibration frequency component in the full frequency band.

4. The electromagnetic force weighing dynamic calibration method of claim 3, wherein, According to the amplitude-frequency characteristic variation of the electromagnetic force, the nonlinear distortion parameters corresponding to each vibration frequency band are extracted to form a comprehensive compensation mechanism, including: Based on the vibration frequency-electromagnetic force amplitude response mapping table, the gain variation of the electromagnetic force in the full frequency band is analyzed, and an abnormal vibration frequency band deviating from the linear theoretical value is obtained through preset gain deviation threshold identification. For the abnormal vibration frequency band, the dynamic deviation between the theoretical prediction value and the sensor measured value of the electromagnetic force in the frequency band is calculated to generate a frequency-indexed nonlinear distortion parameter set. The frequency-indexed nonlinear distortion parameter set is divided into a segmented frequency domain compensation rule set according to the abnormal frequency band boundary frequency division rule to form a dynamic force real-time compensation mechanism.

5. The electromagnetic force weighing dynamic calibration method of claim 4, wherein, Three detection points with fixed spatial positions are set on the logistics sorting line, and the vibration characteristics of each detection point are obtained by combining the comprehensive compensation mechanism. Based on the spatial distribution relationship of the three detection points, a dynamic compensation coefficient is generated, including: In the load direction of the weighing station conveyor belt, three vibration sensors are linearly arranged as detection points at a preset fixed interval along the conveyor belt running path. The first detection point is located at the beginning of the conveyor belt, the second detection point is located at the middle of the conveyor belt, and the third detection point is located at the end of the conveyor belt. The frequency spectrum distribution data of the acceleration signal of the conveyor belt at each detection point is collected in real time. The frequency spectrum distribution data of the three detection points and the corresponding spatial coordinate values are combined to obtain rigid vibration component data and flexible deformation component data of the conveyor belt according to a predefined vibration component separation rule. Based on the rigid vibration component data and the flexible deformation component data, the dynamic influence ratio at the weighing position is calculated to generate a set of adaptive compensation coefficients bound to the spatial position.

6. The electromagnetic force weighing dynamic calibration method of claim 5, wherein, According to the real-time vibration characteristics and the dynamic compensation coefficient in the logistics sorting, the matching relationship of the frequency band parameters in the comprehensive compensation mechanism is dynamically adjusted, including: The real-time collected vibration frequency spectrum data is input into the segmented frequency domain compensation unit, and the dynamic compensation coefficient set is called to reconstruct the compensation weight parameters of each frequency band. Based on the reconstructed compensation weight parameters, the proportion of high-frequency vibration energy in the total frequency spectrum is calculated, and a high-frequency compensation intensity enhancement coefficient is generated according to the proportion value. The low-frequency amplitude offset in the compensation weight parameter is extracted, and the high-frequency compensation intensity enhancement coefficient is combined to dynamically correct the frequency band parameters in the compensation unit.

7. The electromagnetic force weighing dynamic calibration method of claim 6, wherein, Based on the adjusted comprehensive compensation mechanism, the electromagnetic coil current is adjusted to achieve dynamic balance between the electromagnetic force and the measured mass, and the weighing accuracy under different working conditions is obtained, including: The updated comprehensive compensation mechanism parameter set is input into the electromagnetic force control unit, and the vibration disturbance force component is analyzed based on the real-time collected vibration frequency spectrum data to generate a set of frequency-discretely-distributed current compensation signals. The compensation amount of each frequency band in the extraction current compensation signal group is superimposed with the reference drive current of the electromagnetic coil by frequency band in amplitude-phase synchronization to generate an anti-vibration synthesis drive current instruction set; The displacement state sequence of the weighing unit is collected by the displacement sensing device while the electromagnetic coil current supply is controlled by the anti-vibration synthesis drive current instruction set; The dynamic stationarity of the displacement state sequence is detected, and when the displacement fluctuation amplitude is within the zeroing tolerance range for a continuous preset period, the mass measurement value after vibration isolation is obtained.

8. An electromagnetic force weighing dynamic calibration system, the system implementing the method of any one of claims 1 to 7, characterized in that, It comprises: A signal acquisition module is used to acquire composite vibration signals and form working condition vibration characteristic data in the process of logistics sorting under the conditions of high-speed motion, frequent start-stop and random vibration of packages; A signal construction module is used to construct random vibration excitation signals covering the full frequency band based on the working condition vibration characteristic data to generate signals simulating the excitation environment of the multi-frequency superimposed electromagnetic force balancing device in the actual working condition; A synchronous acquisition module is used to synchronously acquire the electromagnetic force signal output by the sensor, the real-time acceleration signal of the measured object and the vibration frequency spectrum data during the calibration process of the electromagnetic force weighing sensor based on the random vibration excitation signal, and record the amplitude-frequency characteristic changes of the electromagnetic force under different vibration components; A parameter extraction module is used to extract the nonlinear distortion parameters corresponding to each vibration frequency band according to the amplitude-frequency characteristic changes of the electromagnetic force to form a comprehensive compensation mechanism; A coefficient generation module is used to set three detection points with fixed spatial positions on the logistics sorting line, obtain the vibration characteristics of each detection point in combination with the comprehensive compensation mechanism, generate dynamic compensation coefficients based on the spatial distribution relationship of the three detection points; A dynamic adjustment module is used to dynamically adjust the matching relationship of the parameters of each frequency band in the comprehensive compensation mechanism according to the real-time vibration characteristics and dynamic compensation coefficients in the logistics sorting; A balance adjustment module is used to adjust the electromagnetic coil current based on the adjusted comprehensive compensation mechanism to realize the dynamic balance between the electromagnetic force and the measured mass and obtain the weighing accuracy under different working conditions.

9. A computing device, comprising: It comprises: One or more processors; A storage device is used to store one or more programs, when the one or more programs are executed by the one or more processors, so that the one or more processors implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program which is executed by the processor to implement the method of any one of claims 1 to 7.

Images