RFID-based warehouse material data acquisition method and system

Abstract

The invention relates to the technical field of data processing, in particular to a warehouse material data acquisition method and system based on RFID, and the method comprises the steps: calculating the feature value of each piece of data in a historical temperature data sequence collected at a preset initial frequency, carrying out the differential iteration of a feature value sequence, and in the iteration process, calculating the feature value of each piece of data in the historical temperature data sequence; calculating the stability of each difference result, if the stability of the difference result is greater than a preset threshold value, stopping iteration to obtain a difference order, inputting the characteristic value sequence and the difference order into a pre-trained ARIMA model, and predicting to obtain a target characteristic value sequence of a next sampling period; and calculating the sampling frequency of the next sampling time period based on the initial frequency and the data mean value in the target characteristic value sequence, so as to collect the temperature data of the warehouse materials in the next sampling time period based on the sampling frequency. According to the invention, the sampling frequency of the temperature data of the warehouse materials can be accurately adjusted, and the acquisition efficiency of the temperature data is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to a method and system for acquiring warehouse material data based on RFID. Background Art

[0002] Radio frequency identification technology (RFID) is a contactless automatic identification technology that uses radio frequency signals to achieve contactless identification of targets and data reading and writing. It has the advantages of low cost, high speed, and long identification distance, and is widely used in warehouse material management. In warehouse management, in order to ensure the safety of materials, it is usually necessary to monitor environmental data such as temperature. However, due to the uncertainty of the warehouse environment, the traditional unified sampling frequency method will increase the system burden and affect efficiency, and cannot take into account the contradiction between the precision of monitoring data changes and the dynamic adjustment of sampling frequency. Therefore, a solution that can accurately and flexibly collect data is needed.

[0003] The Autoregressive Moving Average (ARIMA) model is a forecasting tool widely used in time series analysis, suitable for modeling and forecasting time series data. The model uses a set difference order to stabilize the original data, revealing the underlying laws in the data and accurately predicting future values. Therefore, the model can be used to predict the changes in data in the future period, thereby adjusting the data sampling frequency based on the predicted data changes and achieving self-adaptation of the data sampling frequency.

[0004] In the related art, a method for predicting gas consumption based on an ARIMA model is disclosed in a patent application document with publication number CN115659621A. The method includes: step 1: obtaining a basic data set; step 2: processing data extracted from the basic data set to obtain a data set that meets the user's gas consumption specifications; step 3: constructing an ARIMA model of a single user's gas consumption based on the data set obtained in step 2, and predicting the gas consumption of a single user to obtain a prediction result; step 4: constructing a regional user gas consumption model with different regional attributes, predicting the gas consumption of regional users with different regional attributes, and obtaining a prediction result.

[0005] However, in the above scheme, when the ARIMA model is used to determine the forecast results, the difference order is a preset fixed value. For data with large fluctuations, the fixed difference order may cause the forecast results to be overfitted or underfitted, thereby affecting the accuracy of the forecast results. Summary of the invention

[0006] In order to solve the problem that the sampling frequency of data cannot be accurately adjusted due to the poor accuracy of the prediction results determined by the ARIMA model, the present invention provides a warehouse material data acquisition method and system based on RFID.

[0007] According to a first aspect of an embodiment of the present invention, there is provided a method for acquiring warehouse material data based on RFID, comprising:

[0008] At a preset initial frequency, the temperature data of warehouse materials is collected during the current sampling period to obtain a historical temperature data sequence;

[0009] Calculate the characteristic value of each temperature data in the historical temperature data sequence to obtain the characteristic value sequence. The characteristic value reflects the speed of data change in the local range centered on any temperature data.

[0010] Perform differential iteration on the eigenvalue sequence. During the iteration process, calculate the stability of each differential result. If the stability of the differential result is greater than the preset threshold, stop the iteration and get the differential order. The stability is negatively correlated with the proportion of extreme value points in the differential sequence obtained after the differential is completed.

[0011] Input the eigenvalue sequence and the difference order into the pre-trained ARIMA model to predict the eigenvalue sequence of the next sampling period and obtain the target eigenvalue sequence;

[0012] The sampling frequency of the next sampling period is calculated. The sampling frequency is positively correlated with the initial frequency and the normalized value of the average value of all eigenvalues ​​in the target eigenvalue sequence, so as to collect the temperature data of the warehouse materials in the next sampling period based on the sampling frequency.

[0013] The present invention can adaptively adjust the sampling frequency of temperature data according to the real-time changes of warehouse material temperature data. For time periods with fast data changes, a higher sampling frequency can be used to obtain more detailed information; and for time periods with slow data changes, a lower sampling frequency can be used to reduce unnecessary data storage and transmission, thereby improving overall efficiency. And through the data distribution of the fast and slow changes in temperature data, the differential order of the ARIMA prediction model is adaptively adjusted, which effectively improves the fitting accuracy of the model, reduces overfitting and underfitting problems, and ensures the accuracy of the prediction results, thereby ensuring the accuracy of the adaptive adjustment of the temperature data sampling frequency and improving the efficiency of obtaining temperature data of warehouse materials.

[0014] Preferably, calculating the characteristic value of each temperature data in the historical temperature data sequence includes:

[0015] The data with the largest difference from the mean value of the data in the local range of any temperature data in the historical temperature series is used as the reference data;

[0016] The characteristic value of any temperature data satisfies the following relationship:

[0017] ;

[0018] In the formula, For the The characteristic value of the temperature data; For the The variance of the sequence composed of the first-order differences of adjacent data within the local range of the temperature data; For the The variance of the sequence composed of the first-order differences of adjacent data except the reference data within the local range of the temperature data; To preset hyper parameters; is the sampling frequency of the current sampling period; is the absolute value symbol; is a natural exponential function.

[0019] The present invention can appropriately amplify the data changes in a sampling period with a lower sampling frequency, thereby being able to accurately measure the speed of change of temperature data in different sampling periods.

[0020] Preferably, the method for obtaining reference data includes:

[0021] The data with the largest absolute value of the difference between the local range and the data mean of the local range is selected to obtain the reference data.

[0022] The reference data screened out by the present invention is highly likely to be noise data, so the noise interference can be reduced by calculating the changes in the data segments before and after the reference data are removed, so as to ensure the accuracy of the characteristic values ​​of each temperature data determined.

[0023] Preferably, the method for obtaining the first-order difference of adjacent data within a local range includes:

[0024] Starting from the second data in the local range, calculate the difference between any data and the previous data to obtain the first-order difference of all adjacent data in the local range.

[0025] Preferably, the stability of each differential result satisfies the following relationship:

[0026] ;

[0027] In the formula, For the Stationarity of sub-difference results; For the The number of extreme points in the difference sequence obtained after the second difference is completed; For the The number of data points in the difference sequence obtained after the difference is completed.

[0028] The present invention can determine the stationarity of the difference sequence through a simple calculation method, thereby reducing the difficulty of calculating the stationarity of the difference sequence.

[0029] Preferably, the method for obtaining the number of extreme value points in the difference sequence includes:

[0030] The difference method is used to identify the number of extreme points in the difference sequence obtained after each difference is completed.

[0031] Preferably, the sampling frequency of the next sampling period satisfies the following relationship:

[0032] ;

[0033] In the formula, is the sampling frequency of the next sampling period; To preset the initial frequency; is the average value of all eigenvalues ​​in the target eigenvalue sequence; is a natural exponential function.

[0034] The present invention can reduce the amount of calculation and the difficulty of calculating the sampling frequency in the next time period by normalizing the mean of the eigenvalues ​​in the target eigenvalue sequence.

[0035] According to the second aspect of an embodiment of the present invention, a RFID-based warehouse material data acquisition system is provided. The RFID-based warehouse material data acquisition system includes a memory and a processor, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the steps of the first aspect of the embodiment of the present invention.

[0036] The present invention has the following effects:

[0037] 1. The present invention inputs the eigenvalue sequence of the eigenvalue combination of each temperature data in the historical temperature data sequence and the calculated differential order into the pre-trained autoregressive moving average model to predict the eigenvalue sequence of the next sampling period, which can avoid the overfitting and underfitting of the eigenvalue sequence by the model, ensure the accuracy of the determined target eigenvalue sequence, and realize the precise adjustment of the data sampling frequency.

[0038] 2. The present invention adaptively adjusts the sampling frequency of the temperature data of warehouse materials based on the speed of data change. For time periods with faster data changes, a higher sampling frequency is used to obtain more detailed information; and for time periods with slower data changes, a lower sampling frequency is used, thereby reducing unnecessary data storage and transmission, thereby improving the efficiency of obtaining temperature data of warehouse materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0040] Figure 1 This is a schematic diagram of the steps of a method for acquiring warehouse material data based on RFID according to an embodiment of the present invention;

[0041] Figure 2 It is a schematic diagram representing the relative position of the current sampling period and the next sampling period in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0043] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0044] Reference Figure 1 , a warehouse material data acquisition method based on RFID, including steps S1 to S5, specifically as follows:

[0045] S1: Collect the temperature data of warehouse materials within the current sampling period at a preset initial frequency to obtain a historical temperature data sequence.

[0046] Specifically, you can choose to install RFID temperature sensor tags on materials that need to monitor temperature. These tags are integrated with temperature sensors and can collect temperature data of materials in real time. Then you can use RFID readers to receive wireless signals from RFID tags and transmit temperature data to the central server for processing and analysis.

[0047] Furthermore, the receiving frequency of the RFID reader can be set to the initial frequency , and preset the initial frequency =1Hz. Afterwards, the temperature data of the warehouse materials are collected at this frequency during the current sampling period, thereby obtaining a historical temperature data sequence.

[0048] The current sampling period refers to a time period with the current moment as the end node and a preset duration. In this embodiment, the preset duration is 10 minutes.

[0049] S2: Calculate the eigenvalue of each temperature data in the historical temperature data sequence to obtain an eigenvalue sequence. The eigenvalue reflects the speed of data change in a local range centered on any temperature data.

[0050] In an exemplary embodiment of the present invention, the characteristic value refers to data determined based on the speed of data change in a local range of any temperature data. For example, when the data change in a local range of any temperature data is more drastic, the characteristic value of the temperature data is relatively large; on the contrary, if the data change in a local range of any temperature data is relatively slow, the characteristic value of the temperature data is relatively small.

[0051] The local range refers to the area around any temperature data. In this embodiment, =10.

[0052] Optional, if the amount of data on either side of any temperature data in the historical temperature data sequence is less than , then discard the data on that side and compare the temperature data with the data on the other side. The range of data is taken as the local range of the temperature data.

[0053] In an exemplary embodiment of the present invention, the characteristic value of any temperature data in the historical temperature data sequence can be determined by the following steps:

[0054] Step 1: Take the data with the largest difference from the mean value of the data in the local range of any temperature data in the historical temperature series as the reference data;

[0055] It should be noted that in the process of the RFID reader receiving temperature data, there may be noise data in the historical temperature data sequence due to problems with the warehouse environment or the sensor itself, and the existence of noise data will affect the evaluation of the speed of data change. Therefore, when calculating the characteristic values ​​of each temperature data, it is necessary to take into account the influence of noise data to ensure the accuracy of the determined characteristic values.

[0056] Among them, the reference data refers to the data with the greatest possibility of being noise data within the local range of any temperature data. It should be noted that the mean value of the data within the local range can reflect the overall level of the data within the local range, while the noise data usually exhibits outlier characteristics. When a certain data within the local range deviates greatly from the mean value of the data within the local range, the data is likely to be noise data. Therefore, based on this feature, the data that is most likely to be noise data within the local range can be screened out.

[0057] In an exemplary embodiment of the present invention, the determination of reference data within a local range of any temperature data can be achieved by the following steps:

[0058] The data with the largest absolute value of the difference between the local range and the data mean of the local range is selected to obtain the reference data.

[0059] In another embodiment, the reference data in the corresponding local range can also be screened out according to the ratio of the local range of any temperature data to the data mean value in the local range. This embodiment does not specifically limit the method of measuring the difference between each data in the local range and the data mean value.

[0060] Step 2: Calculate the first-order difference of adjacent data within a local range of any temperature data, and the variance of a sequence consisting of the first-order differences of adjacent data before and after removing the reference data within the local range.

[0061] In an exemplary embodiment of the present invention, the determination of the first-order difference of adjacent data within a local range of any temperature data can be achieved by the following steps:

[0062] Starting from the second data in the local range, calculate the difference between any data and the previous data to obtain the first-order difference of all adjacent data in the local range.

[0063] Among them, the first-order difference of adjacent data reflects the change range of the corresponding two adjacent data. The temperature data is recorded as , will Temperature data In the local area The data is recorded as , then the first-order difference of adjacent data in the local range can be expressed as ; In the formula, For the The first-order difference of adjacent data within the local range of temperature data; For the The local range of temperature data +1 data; For the The local range of temperature data data.

[0064] Furthermore, the The sequence composed of the first-order differences of all adjacent data within the local range of a temperature data is recorded as ; In the formula, is the difference between the second data and the first data in the local range, is the difference between the third data and the second data in the local range, is the first Data and The difference between the data, is the number of data in the local range. Similarly, a sequence consisting of first-order differences of adjacent data after removing the reference data in the local range can be obtained, which will not be described in detail in this embodiment.

[0065] Furthermore, after determining a sequence composed of the first-order differences of adjacent data before and after removing the reference data in a local range of any temperature data, the variance of the two sequences can be obtained. It should be noted that the calculation method of the variance of the data sequence is a prior art and is not described in detail in this embodiment.

[0066] Step 3: Use the first variance to weight the second variance, and calculate the characteristic value of the temperature data by combining the weighted value and the sampling frequency of the current sampling period.

[0067] Specifically, the characteristic value of any temperature data in the historical temperature data sequence satisfies the following relationship:

[0068] ;

[0069] In the formula, For the The characteristic value of the temperature data; For the The variance of the sequence composed of the first-order differences of adjacent data within the local range of the temperature data; For the The variance of the sequence composed of the first-order differences of adjacent data except the reference data within the local range of the temperature data; To preset hyperparameters, this embodiment , in order to prevent the denominator of the eigenvalue calculation formula from being zero; is the sampling frequency of the current sampling period; is the absolute value symbol; is the natural exponential function, where the natural exponential function is the natural constant An exponential function with base .

[0070] in, It can characterize the consistency of data changes within the corresponding local range. When the value is larger, the data in the corresponding local range changes more dramatically, and the corresponding local range is more likely to contain more detailed information; on the contrary, when The smaller it is, the slower the data changes in the corresponding local range, and the less detailed information there is in the corresponding local range.

[0071] It reflects the influence of the reference data in the local range of any temperature data on the data change in the local range. The larger the value, the greater the influence of the reference data on the data change in the local range, which further indicates that the reference data in the local range is more likely to be noise data. When it is large, that is, when the data in the local range changes drastically, it will lead to The larger the reason, the more likely it is caused by noise data. and The product of The value of The value of The credibility of the value is to reduce the impact of noisy data and achieve accurate measurement of data changes in a local area.

[0072] It should be noted that, since the present invention needs to adjust the sampling frequency of the temperature data of the next sampling period according to the change of the temperature data of any sampling period, the sampling frequencies of different sampling periods are different, that is, the time intervals of adjacent temperature data of different sampling periods are different, which affects the accuracy of the characteristic value. Therefore, the present invention introduces a compensation factor when calculating the characteristic value of temperature data, that is, , so that the change of temperature data in the sampling period with lower sampling frequency can be magnified, and the change of temperature data in the sampling period with higher sampling frequency can be reduced, so as to reduce the impact of different sampling frequencies and ensure the accuracy of the characteristic values ​​of temperature data in each sampling period.

[0073] S3: Perform differential iteration on the eigenvalue sequence. During the iteration process, calculate the stationarity of each differential result. If the stationarity of the differential result is greater than a preset threshold, stop the iteration and obtain the differential order.

[0074] It should be noted that the Autoregressive Integrated Moving Average (ARIMA) model can predict the change pattern of data in the next sampling period over time based on the change pattern of data in the current sampling period over time. Therefore, based on the change relationship of the characteristic value of each temperature data in the current sampling period over time, the model can be used to predict the change relationship of the characteristic value of each temperature data in the next sampling period over time. Then, according to the predicted change relationship of the characteristic value of the next sampling period over time, the frequency of collecting temperature data of warehouse materials in the next sampling period can be adjusted.

[0075] It should be further explained that in the ARIMA model, the purpose of the difference order is to convert a non-stationary time series into a stationary time series; the essence of the difference is to start from the second data in the sequence and subtract the previous data in sequence, so as to make the sequence stationary; the difference order and the number of differences are in a one-to-one correspondence. Therefore, in order to improve the problem of overfitting and underfitting of the sequence caused by the traditional ARIMA model using a fixed difference order, the present invention improves the traditional ARIMA model, and the specific improvement is: firstly, the sequence is differentiated, and the stationarity of each difference result is calculated, so as to determine the final difference order based on the set termination condition, so as to realize the self-adaptation of the difference order, and ensure the accuracy of the prediction results obtained based on the improved ARIMA model.

[0076] Specifically, in the process of iteratively differencing the eigenvalue sequence, the stability of any difference result can be determined by the following steps:

[0077] Step 1: Calculate the proportion of extreme value points in the difference sequence obtained after any difference is completed;

[0078] First, we can determine the difference sequence obtained after each difference is completed. Specifically, the eigenvalue sequence can be recorded as ,in, is the characteristic value of the first temperature data in the historical temperature data sequence, is the characteristic value of the second temperature data in the historical temperature data sequence, The historical temperature data series The characteristic value of the temperature data, is the number of data points in the eigenvalue sequence; then the eigenvalue sequence The difference sequence after the first difference is completed can be recorded as , the difference sequence after the second difference is completed can be recorded as , and so on, The difference sequence after the second difference is completed can be recorded as , that is, after each difference is completed, the number of data points in the eigenvalue sequence is reduced by 1, so that the difference sequence obtained after each difference is completed can be obtained.

[0079] After that, it is necessary to identify the extreme points in the difference sequence obtained after each difference is completed, and then calculate the proportion of extreme points in the difference sequence obtained after each difference is completed.

[0080] In an exemplary embodiment of the present invention, the number of extreme value points in the differential sequence obtained after any difference is completed can be determined by the following steps:

[0081] The difference method is used to identify the number of extreme points in the difference sequence obtained after each difference is completed.

[0082] Optionally, other extreme point identification methods may be used, such as local maximum and local minimum detection methods, peak detection algorithms, etc. This embodiment does not specifically limit the method used to identify extreme points. It should be noted that using the differential method to identify extreme points in a data sequence is a prior art, and this embodiment will not be described in detail.

[0083] Step 2: Based on the proportion of extreme points in the differential sequence obtained after the difference is completed, calculate the stationarity of the difference result. The stationarity is negatively correlated with the proportion.

[0084] Specifically, the stability of any difference result satisfies the following relationship:

[0085] ;

[0086] In the formula, For the Stationarity of sub-difference results; For the The number of extreme points in the difference sequence obtained after the second difference is completed; For the The number of data points in the difference sequence obtained after the difference is completed.

[0087] in, Reflects the eigenvalue sequence in The proportion of extreme points in the difference sequence obtained after the sub-difference is completed. The larger the value, the greater the volatility of the difference sequence, and the corresponding sub-difference result has poorer stability.

[0088] In another embodiment, the stationarity of any differential result can also be calculated by other calculation formulas, such as not determining it by a calculation formula that differs 1 from the extreme point ratio, but by using the opposite of the extreme point ratio as the power of a natural exponential function to perform a power operation.

[0089] Further, after calculating the stability of any difference result, the stability of the difference result can be compared with a preset threshold value. In this embodiment, the preset threshold value is 0.83. If the stability of any difference result is greater than 0.83, it means that the difference sequence obtained by the difference is relatively stable, then the difference iteration process is stopped, and the current number of differences is used as the difference order. Of course, a suitable threshold value can also be set according to the specific situation, and this embodiment does not specifically limit the setting of the threshold value.

[0090] S4: Input the eigenvalue sequence and the difference order into the pre-trained ARIMA model to predict the eigenvalue sequence of the next sampling period and obtain the target eigenvalue sequence.

[0091] refer to Figure 2 As shown, if the current sampling period is When , the next sampling period is , if the current sampling period is , the next sampling period is , and the duration of each sampling period is the same, all 10 minutes.

[0092] Optionally, before the eigenvalue sequence and the difference order are input into the pre-trained autoregressive moving average model, the ARIMA model needs to be trained. The sample set used in the training process is an eigenvalue sequence composed of the eigenvalues ​​of the temperature data in several sampling periods, the difference order is the difference order determined by step S3, and the minimization loss function used can be to minimize the residual sum of squares. It should be noted that the training process of the ARIMA model is a prior art, and this embodiment will not be described in detail here.

[0093] Optionally, the present invention uses the ARIMA model to predict the eigenvalue sequence of the next sampling period based on the adaptively determined difference order, which can avoid the overfitting and underfitting of the eigenvalue sequence by the ARIMA model and ensure the accuracy of the obtained target eigenvalue sequence.

[0094] S5: Calculate the sampling frequency of the next sampling period. The sampling frequency is positively correlated with the initial frequency and the normalized value of the average value of all eigenvalues ​​in the target eigenvalue sequence, so as to collect the temperature data of the warehouse materials in the next sampling period based on the sampling frequency.

[0095] Specifically, the sampling frequency of the next sampling period satisfies the following relationship:

[0096] ;

[0097] In the formula, is the sampling frequency of the next sampling period; To preset the initial frequency, in this embodiment =1Hz, it should be noted that the initial frequency set in the present invention is relatively large, therefore, the present invention only adjusts the initial frequency downward; is the average value of all eigenvalues ​​in the target eigenvalue sequence; is a natural exponential function.

[0098] In another embodiment, it is also possible to use Function, normalizes the average values ​​of all eigenvalues ​​in the target eigenvalue sequence. This embodiment does not specifically limit the selected normalization method.

[0099] In another embodiment, the sampling frequency of the next sampling period may also be calculated by other calculation formulas, such as by a summation formula instead of a product formula.

[0100] The present invention can accurately adjust downward the initial frequency of the temperature data of warehouse materials, while ensuring that the collected temperature data contains sufficient detail information, it can reduce the storage and transmission of unnecessary data, thereby improving the efficiency of obtaining the temperature data of warehouse materials.

[0101] The present invention also provides an RFID-based warehouse material data acquisition system, the system includes a memory and a processor, and a computer program is stored in the memory, the computer program integrates a RFID-based warehouse material data acquisition method function, when the computer program is executed, a RFID-based warehouse material data acquisition method can be used to accurately adjust the sampling frequency of the temperature data of the warehouse materials, thereby improving the efficiency of obtaining the temperature data.

[0102] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0103] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A warehouse material data acquisition method based on RFID, characterized in that: include: At a preset initial frequency, the temperature data of warehouse materials is collected during the current sampling period to obtain a historical temperature data sequence; Calculate the characteristic value of each temperature data in the historical temperature data sequence to obtain a characteristic value sequence, wherein the characteristic value reflects the speed of data change in a local range centered on any temperature data; Performing differential iteration on the eigenvalue sequence, during the iteration process, calculating the stationarity of each differential result, if the stationarity of the differential result is greater than a preset threshold, stopping the iteration, and obtaining the differential order; The stability is negatively correlated with the proportion of extreme value points in the difference sequence obtained after the difference is completed; Input the eigenvalue sequence and the difference order into the pre-trained ARIMA model to predict the eigenvalue sequence of the next sampling period to obtain the target eigenvalue sequence; The sampling frequency of the next sampling period is calculated, and the sampling frequency is positively correlated with the initial frequency and the normalized value of the average value of all eigenvalues ​​in the target eigenvalue sequence, so as to collect the temperature data of the warehouse materials in the next sampling period based on the sampling frequency.

2. The method for acquiring warehouse material data based on RFID according to claim 1, characterized in that: The calculating the characteristic value of each temperature data in the historical temperature data sequence includes: The data with the largest difference from the mean value of the data in the local range of any temperature data in the historical temperature sequence is used as the reference data; The characteristic value of any temperature data satisfies the following relationship: ； In the formula, For the The characteristic value of the temperature data; For the The variance of the sequence composed of the first-order differences of adjacent data within the local range of the temperature data; For the The variance of the sequence composed of the first-order differences of adjacent data except the reference data within the local range of the temperature data; To preset hyper parameters; is the sampling frequency of the current sampling period; is the absolute value symbol; is a natural exponential function.

3. The method for acquiring warehouse material data based on RFID according to claim 2, characterized in that: The method for obtaining the reference data includes: The data having the largest absolute value of the difference between the local range and the data mean of the local range is selected to obtain the reference data.

4. The method for acquiring warehouse material data based on RFID according to claim 2, characterized in that: The method for obtaining the first-order difference of adjacent data within the local range includes: Starting from the second data in the local range, the difference between any data and the previous data is calculated to obtain the first-order difference of all adjacent data in the local range.

5. The method for acquiring warehouse material data based on RFID according to claim 1, characterized in that: The stability of each difference result satisfies the following relationship: ； In the formula, For the Stationarity of sub-difference results; For the The number of extreme points in the difference sequence obtained after the second difference is completed; For the The number of data points in the difference sequence obtained after the difference is completed.

6. The method for acquiring warehouse material data based on RFID according to claim 5, characterized in that: The method for obtaining the number of extreme value points in the difference sequence includes: The difference method is used to identify the number of extreme points in the difference sequence obtained after each difference is completed.

7. The method for acquiring warehouse material data based on RFID according to claim 1, characterized in that: The sampling frequency of the next sampling period satisfies the following relationship: ； In the formula, is the sampling frequency of the next sampling period; To preset the initial frequency; is the average value of all eigenvalues ​​in the target eigenvalue sequence; is a natural exponential function.

8. A warehouse material data acquisition system based on RFID, characterized in that: The RFID-based warehouse material data acquisition system includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the steps of the RFID-based warehouse material data acquisition method as described in any one of claims 1-7.

Images