A trend correction method and system based on loading equipment sensing data and a storage medium

Abstract

This invention discloses a trend correction method, system, and storage medium based on sensor data from a loading device, comprising the following steps: periodically acquiring raw data and generating corresponding data groups to be processed; analyzing the current behavioral trend of the data groups to be processed and analyzing the recent behavioral trend of historical data groups in a preset second unit; combining the current behavioral trend and the recent behavioral trend into a short-term behavioral trend, and determining whether a special behavioral trend identical to the short-term behavioral trend exists in a preset comparison database; if a special behavioral trend identical to the short-term behavioral trend exists in the preset comparison database, then supplementing the data group to be processed with the earliest storage time in the first unit, and storing the supplemented data group to be processed as a historical data group. Supplementing data only for data groups deemed important can not only reduce the amount of data processing but also reduce the omission of important data.

Description

Technical Field

[0001] This invention relates to the field of data processing, and in particular to a trend correction method, system, and storage medium based on sensor data from loading equipment. Background Technology

[0002] Warehouses typically use loading equipment to move goods, enabling them to be stored in and out of the warehouse.

[0003] Of course, to enable back-end staff to understand the real-time status of the loading equipment, several sensors need to be installed on the equipment, commonly including positioning sensors and pressure sensors. During operation, the positioning and pressure sensors need to periodically transmit data to the back-end system to provide feedback to the staff regarding the location of the loading equipment and the weight of the goods being transported. Based on this feedback data, the back-end system can also construct a virtual 3D warehouse, displaying the status of goods and loading equipment within the warehouse, facilitating staff's understanding of the storage situation and scheduling.

[0004] In practical use, because the loading equipment needs to be moved, data transmission is generally conducted wirelessly. However, external interference during data transmission can easily lead to data loss, preventing some data from being transmitted to the backend. To ensure that the data presented to staff is more comprehensive and accurate, it is necessary to correct any missing data.

[0005] However, only data from short periods such as lifting and lowering cargo, and slowing down are critical. The loss of individual data during other times will not affect the overall situation. Therefore, a trend correction method based on the sensor data of the loading equipment is needed. Summary of the Invention

[0006] This application provides a trend correction method, system, and storage medium based on loading equipment sensor data, which supplements only the important data and reduces the amount of computation in data processing.

[0007] Firstly, this application provides a trend correction method based on sensor data from loading equipment, employing the following technical solution:

[0008] A trend correction method based on loading equipment sensor data includes the following steps:

[0009] Periodically acquire raw data, and generate corresponding data groups to be processed based on the raw data when the amount of raw data reaches a first preset value;

[0010] The generated data groups to be processed are stored in a preset first unit, and the data groups to be processed in the first unit are arranged in the order of storage.

[0011] Whenever the number of data groups to be processed in the first unit exceeds the second preset value, the current behavior trend of the data groups to be processed in the first unit is analyzed, and the recent behavior trend of the historical data groups in the preset second unit is also analyzed.

[0012] The current behavioral trend and recent behavioral trend are combined into a short-term behavioral trend, and it is determined whether there is a special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend. The comparison database stores multiple special behavioral trends.

[0013] If it is determined that there is a special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend, then the data group to be processed with the earliest storage time in the first unit is supplemented, and the supplemented data group to be processed is stored as a historical data group.

[0014] If it is determined that there is no special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend, then the data group to be processed with the earliest storage time in the first unit is directly stored as the historical data group.

[0015] Preferably, the raw data includes location information and weight information, and the step of generating a corresponding data set to be processed based on the raw data includes the following steps:

[0016] Generate a blank set, the blank set including a number of empty first reserved bits and empty second reserved bits, wherein each first reserved bit is used to store a location information and each second reserved bit is used to store a weight information;

[0017] The corresponding location information is stored sequentially into multiple first reserved bits according to the order in which the original data was acquired.

[0018] The corresponding weight information is stored sequentially into multiple second reserved bits according to the order in which the original data was acquired.

[0019] The blank set containing the corresponding location and weight information in both the first and second reserved bits is converted into a data group to be processed.

[0020] Preferably, before storing the corresponding location information into multiple first reserved bits in chronological order of the original data acquisition time, the method further includes the following steps:

[0021] Determine whether the location information is within a preset first range.

[0022] If the location information is within the preset first range, the location information is retained;

[0023] If the location information is not within the preset first range, then replace the corresponding location information with a null value;

[0024] Before storing the corresponding weight information into multiple second reserved bits in chronological order of the original data acquisition time, the method further includes the following steps:

[0025] Determine whether the weight information is within a preset second range.

[0026] If the weight information is within the preset second range, the weight information is retained;

[0027] If the weight information is not within the preset second range, the corresponding position information will be replaced with a null value.

[0028] Preferably, supplementing the data group to be processed in the first unit with the oldest storage time includes the following steps:

[0029] Determine whether there is a null value in the first reserved bit and whether there is a null value in the second reserved bit.

[0030] If there is a null value in the first reserved position, a weight estimate is generated according to the preset first supplementation rule to replace the null value in the first reserved position;

[0031] If there is a vacancy in the second reserved position, a position estimate is generated according to the preset second supplementation rule to replace the vacancy in the second reserved position;

[0032] If neither the first nor the second reserved bit contains a null value, the data group to be processed will be stored as a historical data group.

[0033] Preferably, the first supplementation rule includes: defining consecutive null values ​​as an empty set, using the weight information at both ends of each empty set as the weight reference information of the corresponding empty set according to the storage time order, and replacing the null values ​​in the empty set with two corresponding weight reference information, wherein each null value is preferentially replaced with weight reference information stored at a similar time.

[0034] Preferably, the second supplementation rule includes: defining consecutive null values ​​as an empty set, using the position information at both ends of each empty set as position reference information for the corresponding empty set according to the storage time order, generating position prediction information based on the two corresponding position reference information, and replacing the null values ​​in the corresponding empty set with the position prediction information.

[0035] Preferably, the step of generating location estimation information based on two corresponding location reference information includes the following steps:

[0036] Determine whether the distance between two corresponding location references exceeds a third preset value.

[0037] If the distance between two corresponding location references exceeds the third preset value, then the same number of location prediction information as the number of null values ​​in the corresponding empty set is generated based on the two corresponding location references, and the location prediction information replaces the null values ​​in the corresponding empty set in turn.

[0038] If the distance between two corresponding location references is less than or equal to a third preset value, then the average value of the two corresponding location references is calculated and used as the location prediction information, and the location prediction information replaces all null values ​​in the corresponding empty set.

[0039] Secondly, this application provides a trend correction system based on sensor data of loading equipment, which adopts the following technical solution:

[0040] A trend correction system based on loading equipment sensor data includes:

[0041] The data acquisition module is used to periodically acquire raw data and generate corresponding data groups to be processed based on the raw data when the amount of raw data reaches a first preset value.

[0042] The data storage module is used to store the generated data to be processed into a preset first unit and to arrange the data to be processed in the first unit in the order of storage.

[0043] The data analysis module is used to analyze the current behavior trend of the data groups to be processed in the first unit whenever the number of data groups to be processed in the first unit exceeds the second preset value, and to analyze the recent behavior trend of the historical data groups in the preset second unit.

[0044] The data comparison module is used to combine the current behavior trend and the recent behavior trend into a short-term behavior trend, and to determine whether there is a special behavior trend in the preset comparison database that is the same as the short-term behavior trend. The comparison database stores multiple special behavior trends.

[0045] If it is determined that there is a special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend, then the data group to be processed with the earliest storage time in the first unit is supplemented, and the supplemented data group to be processed is stored as a historical data group.

[0046] If it is determined that there is no special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend, then the data group to be processed with the earliest storage time in the first unit is directly stored as the historical data group.

[0047] Thirdly, this application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed by the aforementioned trend correction method based on sensor data from a loading device.

[0048] In summary, assessing the importance of each data set to be processed and supplementing the data sets deemed important can not only reduce the amount of data processing but also minimize the omission of important data. Attached Figure Description

[0049] Figure 1 This is an overall flowchart of an embodiment of this application. Detailed Implementation

[0050] The following is in conjunction with the appendix Figure 1 This application will be described in further detail.

[0051] A trend correction method based on sensor data from loading equipment is applied in a processor. This processor is used in the control system of an automated warehouse. The automated warehouse is equipped with several loading devices for handling goods. Each loading device is equipped with a data acquisition device, which includes a pressure sensor and a positioning sensor. The processor is used to process the data collected by the data acquisition device.

[0052] A trend correction method based on loading equipment sensor data, see [link to relevant documentation]. Figure 1 This includes the following steps:

[0053] S100. Periodically acquire raw data, and generate corresponding data groups to be processed based on the raw data when the amount of raw data reaches a first preset value.

[0054] The first preset value is a manually set value; in this embodiment, the first preset value is 5. Correspondingly, the data group to be processed stores 5 original data.

[0055] Raw data refers to the data collected by the acquisition devices on the loading equipment. Pressure sensors detect the weight of the goods being handled by the loading equipment and output corresponding weight information, while positioning sensors detect the position of the loading equipment and output corresponding positioning information. Therefore, the raw data includes both weight and positioning information. In other words, the data set to be processed actually stores five weight information entries and five positioning information entries. To facilitate data storage, the data set includes a first reserved space for storing positioning information and a second reserved space for storing weight information. Weight information represents the weight of the goods and is a numerical value, while positioning information represents the position of the loading equipment and is a coordinate system. In automated warehouses, it is generally not necessary to know the vertical position of the loading equipment; therefore, the positioning information only needs to be usable in a planar coordinate system. In this embodiment, the positioning information represents the coordinate point (X, Y).

[0056] In addition, due to data loss, the processor may fail to receive weight information and / or location information within the specified time interval. In such cases, the processor will generate weight information and / or location information representing a null value after the specified time interval has ended, so as to facilitate subsequent steps.

[0057] The need to periodically acquire raw data stems from the processor's limited data processing capabilities, preventing it from receiving and processing data continuously. Therefore, a specific interval needs to be set for data acquisition; in this implementation, the interval is 200ms. This means the processor acquires raw data every 200ms.

[0058] It should be noted that the timing interval can be set either in the processor or on the pressure sensor and positioning sensor. That is, the method for periodically acquiring raw data can be either for the processor to receive raw data every 200ms, or for the pressure sensor and positioning sensor to transmit corresponding data to the processor every 200ms.

[0059] S200: Store the generated data to be processed into a preset first unit, and arrange the data to be processed in the first unit in the order of storage.

[0060] The first unit is a virtual storage space. To facilitate the arrangement of data groups to be processed in the order of storage priority, a storage time is generated for each data group to be processed when it is stored in the first unit, and a correspondence is established between the storage time and the data group to be processed. The storage time is also stored in the first unit.

[0061] S300. Whenever the number of data groups to be processed in the first unit exceeds the second preset value, analyze the current behavior trend of the data groups to be processed in the first unit, and analyze the recent behavior trend of the historical data groups in the preset second unit.

[0062] The second preset value is also manually set; in this embodiment, the second preset value is 2. That is, when the first unit stores 3 groups of data to be processed, the analysis of current and recent behavioral trends will be initiated.

[0063] The second unit is also a virtual storage space. Historical data groups are transformed from the data groups to be processed through subsequent steps. Compared to the data groups to be processed, historical data groups have higher accuracy, therefore the recent behavioral trends formed by historical data groups are more accurate than the current behavioral trends. The number of historical data groups stored in the second unit is greater than the number of data groups to be processed stored in the first unit, which makes recent behavioral trends more pronounced than current behavioral trends. Of course, to reduce computational pressure, the second unit does not store an unlimited number of historical data groups; the number is limited, for example, manually set to store only 10 historical data groups. When the number of historical data groups in the second unit reaches this limit, if a new historical data group is generated, the oldest historical data group stored in the second unit is deleted, and the new historical data group is added to the second unit.

[0064] S400: Combine the current behavioral trend and the recent behavioral trend into a short-term behavioral trend, and determine whether there is a special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend.

[0065] The comparison database is a pre-set database used to store multiple special behavioral trends, and the special behavioral trends stored in the comparison database are manually entered.

[0066] Special behavioral trends refer to the data change trends of loading equipment during important operations. Important operations are those that cause changes in the operation of the loading equipment. The loading equipment can be a crane or a forklift, as long as it can move goods. In this embodiment, a forklift is used as an example. Important operations include the brief process of the forklift moving from a stopped state to forward movement, the brief process of the forklift lifting the goods from initial contact, and the brief process of the forklift lowering and separating the goods. Each important operation has a corresponding behavioral trend, and these behavioral trends are recorded in a comparison database to form special behavioral trends.

[0067] Since the data collection times of the current behavioral trend and the historical data group corresponding to the recent behavioral trend are adjacent, the method of combining the current behavioral trend and the recent behavioral trend into a short-term behavioral trend can be simply to directly concatenate the recent behavioral trend and the current behavioral trend according to time order.

[0068] S500. If it is determined that there is a special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend, then the data group to be processed with the earliest storage time in the first unit is supplemented, and the supplemented data group to be processed is stored as a historical data group.

[0069] Due to the inaccuracy of the data sets corresponding to the current behavioral trend, the resulting short-term behavioral trend also exhibits significant inaccuracy, making it difficult to achieve a 100% match rate between short-term and special behavioral trends. Therefore, in determining whether a special behavioral trend identical to the short-term behavioral trend exists in the preset comparison database, a match rate not lower than a preset baseline value is sufficient. This baseline value can be set manually or by comparing the errors between short-term and special behavioral trends corresponding to various manually identified important operations. The maximum error can be analyzed, and the baseline value can be set based on this maximum error to reduce misjudgments of short-term behavioral trends corresponding to important operations.

[0070] Supplementation refers to adding numerical values ​​to the weight and / or location information that represent null values ​​in the data set to be processed, so as to make the entire data set complete.

[0071] In one embodiment, supplementing the data group to be processed in the first unit with the oldest storage time includes the following steps:

[0072] S510. Determine whether there is a null value in the first reserved bit and whether there is a null value in the second reserved bit.

[0073] S520. If there is a null value in the first reserved position, generate a weight estimate according to the preset first supplementation rule to replace the null value in the first reserved position.

[0074] S530. If there is a vacancy in the second reserved position, generate a position estimate according to the preset second supplementation rule to replace the vacancy in the second reserved position.

[0075] S540. If there are no null values ​​in either the first reserved bit or the second reserved bit, then the data group to be processed is stored as a historical data group.

[0076] If any null value exists in the data group to be processed, regardless of whether it is in the first or second reserved position, the null value needs to be replaced with actual data. However, if there are no null values ​​in the data group to be processed, there is no need to pad the data group, so it can be directly treated as a historical data group.

[0077] Furthermore, the first supplementation rule includes: defining consecutive null values ​​as an empty set, using the weight information at both ends of each empty set as the weight reference information of the corresponding empty set according to the storage time order, and using the two corresponding weight reference information to replace the null values ​​in the empty set, wherein each null value is preferentially replaced with the weight reference information stored at a similar time.

[0078] Consecutive null values ​​refer to null values ​​stored at adjacent times. If there is specific data before and after the storage time corresponding to a null value, then this single null value will also be defined as an empty set. That is, the number of null values ​​in the empty set is at least 1, and the maximum number is the same as the first preset value.

[0079] The weight information located at both ends of each empty set according to the storage time order may be the weight information in the same data group to be processed, the weight information with the earliest storage time in the adjacent data group to be processed, or the weight information with the latest storage time in the adjacent historical data group.

[0080] For example, if the data in the current pending data group is stored chronologically as A, empty, empty, empty, B, where A and B are actual data, then there is only one empty set, and the weight reference information is A and B. If the data in the current pending data group is stored chronologically as empty, empty, empty, A, B, and the adjacent historical data group is C, D, E, F, G, there is still only one empty set, but G and A are the weight reference information for that empty set. If the data in the current pending data group is stored chronologically as A, empty, B, empty, empty, and the adjacent pending data group is empty, C, D, E, F, then there are two empty sets corresponding to the current pending data. The weight reference information for the first empty set is A and B, and the weight reference information for the second empty set is B and C.

[0081] After obtaining the weight reference information, all null values ​​in the empty set will be replaced with the weight reference information, prioritizing the replacement of null values ​​with weight reference information from the nearest side. For example, if the data stored in the current data group to be processed in chronological order are A, null, null, B, and C, and the storage time corresponding to the previous null value is adjacent to the storage time corresponding to A, then the previous null value will be replaced with A; if the storage time corresponding to the next null value is adjacent to the storage time corresponding to B, then the next null value will be replaced with B.

[0082] Furthermore, when the number of null values ​​between two corresponding weight references is odd, the null value in the middle will be randomly replaced by one of the two weight references.

[0083] Furthermore, the second supplementary rule includes: defining consecutive null values ​​as an empty set, using the position information at both ends of each empty set as position reference information for the corresponding empty set according to the storage time order, generating position prediction information based on the two corresponding position reference information, and using the position prediction information to replace the null values ​​in the corresponding empty set.

[0084] The method for determining position reference information is the same as that for determining weight reference information, and will not be repeated here.

[0085] However, it can be seen that there is a difference between supplementing weight information and supplementing position information. That is, after obtaining position reference information, the position reference information is not directly used to replace null values. Instead, position prediction information is generated and then used to replace the corresponding null values.

[0086] This is because weight information can change abruptly. For example, when a forklift picks up or puts down a load, there is a noticeable change in weight in an instant. When a forklift starts or stops, it needs to accelerate or decelerate, and the speed changes gradually, so the position does not suddenly stop.

[0087] In one embodiment, generating location prediction information based on two corresponding location references includes the following steps:

[0088] S531. Determine whether the distance between two corresponding position reference information exceeds a third preset value.

[0089] S532. If the distance between two corresponding location reference information exceeds a third preset value, then generate a number of location prediction information that is the same as the number of null values ​​in the corresponding empty set based on the two corresponding location reference information, and replace the null values ​​in the corresponding empty set with the location prediction information in turn.

[0090] S533. If the distance between two corresponding location reference information is less than or equal to a third preset value, then calculate the average value of the two corresponding location reference information and use it as the location prediction information, and replace all null values ​​in the corresponding empty set with the location prediction information.

[0091] The third preset value is set manually. The distance between two corresponding position reference information refers to the distance between the coordinate points represented by the two position reference information in the plane coordinate system.

[0092] When the distance between two location references exceeds a third preset value, it indicates that the loading device has moved a significant distance during this period. Therefore, the location should differ significantly across time intervals, necessitating the generation of multiple different location predictions. The method for generating multiple location predictions is to take points at equal intervals between the two location references. For example, if the coordinates of the two location references are (1,1) and (4,4), and the empty set between them contains two null values, then the corresponding coordinates of the location predictions would be (2,2) and (3,3).

[0093] When the distance between two location references is less than or equal to the third preset value, it indicates that the loading device has moved a small distance during this period and may not be moving at all. It is only necessary to take the average value. For example, if the coordinates of the two location references are (1,1) and (4,4) respectively, and the empty set between them contains two empty values, then the coordinates of the corresponding location prediction information are (2.5,2.5).

[0094] S600. If it is determined that there is no special behavioral trend in the preset comparison database that is the same as the short-term behavioral trend, then the data group to be processed with the earliest storage time in the first unit is directly stored as the historical data group.

[0095] If it is determined that there is no special behavior trend in the preset comparison database that is the same as the short-term behavior trend, it means that the loading device corresponding to the short-term behavior trend is performing routine operation. In this case, the corresponding data group to be processed is not important. Therefore, the data group to be processed with the earliest storage time in the first unit can be directly stored as the historical data group.

[0096] In one embodiment, generating a corresponding group of data to be processed based on the original data includes the following steps:

[0097] S610, Generate an empty set.

[0098] The blank set includes a first reserved space and a second reserved space, and the number of both the first reserved space and the second reserved space is a first preset value. However, both the first reserved space and the second reserved space in the blank set are empty.

[0099] S620. Store the corresponding location information into multiple first reserved bits in the order of the original data acquisition time.

[0100] S630. Store the corresponding weight information into multiple second reserved bits in the order of the original data acquisition time.

[0101] S640. Convert the blank set where both the first and second reserved bits store corresponding position and weight information into a data group to be processed.

[0102] Besides the fact that raw data containing null values ​​can affect the judgment of important operations, the data acquisition device may also make errors when generating raw data, resulting in abnormal data with values ​​that are significantly too large or too small. If these raw data are used directly, they will greatly affect the short-term behavioral trends generated subsequently, leading to inaccurate judgment results. Therefore, it is necessary to filter out the abnormal data before using the raw data.

[0103] Therefore, before storing the corresponding location information into multiple first reserved bits in chronological order of the original data acquisition time, the following steps are also included:

[0104] S621. Determine whether the location information is within the preset first range.

[0105] S622. If the location information is within the preset first range, then the location information is retained.

[0106] S623. If the location information is not within the preset first range, then replace the corresponding location information with a null value.

[0107] The first range is the coordinate range of the area accessible to the loading equipment within the automated warehouse, as defined in the coordinate system. Theoretically, as long as the loading equipment remains within the automated warehouse, the location information uploaded by the loading equipment should fall within this first range. If the location information is not within the preset first range, it is highly likely that the positioning sensor is malfunctioning, and the positioning information it collects is inaccurate. Therefore, the value of the positioning information needs to be reset to empty.

[0108] Accordingly, before storing the corresponding weight information into multiple second reserved bits in chronological order of the original data acquisition time, the following steps are also included:

[0109] S631. Determine whether the weight information is within the preset second range.

[0110] S632. If the weight information is within the preset second range, then the weight information is retained.

[0111] S633. If the weight information is not within the preset second range, the corresponding position information is replaced with a null value.

[0112] The second range refers to the range of cargo weights that the loading equipment can carry during normal use.

[0113] Similar to positioning information, weight information will normally only fall within the preset second range. If the weight information is not within the second range, it is considered that the pressure sensor is malfunctioning, the weight information is inaccurate, and it needs to be replaced with a null value.

[0114] In addition, all raw data is synchronized to the cloud database before being used to determine whether it falls within the preset first and second ranges, thus preserving the unprocessed data. Because the cloud database has greater storage capacity, it can store more raw data, making it suitable for analyzing long-term and continuous raw data. This facilitates the timely detection of any anomalies in the data acquisition equipment. For example, if the weight information in the raw data remains null for 5 minutes, the corresponding pressure sensor is considered to have a potential anomaly such as a power outage; if the position signal in the raw data frequently fluctuates between the first and second ranges within 3 minutes, the corresponding positioning sensor is considered to have a potential anomaly such as malfunction.

[0115] A trend correction system based on sensor data from loading equipment includes a data acquisition module, a data storage module, a data analysis module, and a data comparison module.

[0116] The data acquisition module is used to periodically acquire raw data and generate corresponding data groups to be processed based on the raw data when the amount of raw data reaches a first preset value.

[0117] The data storage module is used to store the generated data to be processed into a preset first unit and to arrange the data to be processed in the first unit in the order of storage.

[0118] The data analysis module is used to analyze the current behavior trend of the data groups to be processed in the first unit whenever the number of data groups to be processed in the first unit exceeds a second preset value, and to analyze the recent behavior trend of the historical data groups in the preset second unit.

[0119] The data comparison module is used to combine the current behavior trend and the recent behavior trend into a short-term behavior trend, and to determine whether there is a special behavior trend in the preset comparison database that is the same as the short-term behavior trend. The comparison database stores multiple special behavior trends. If it is determined that there is a special behavior trend in the preset comparison database that is the same as the short-term behavior trend, the data group to be processed with the earliest storage time in the first unit is supplemented, and the supplemented data group to be processed is stored as a historical data group. If it is determined that there is no special behavior trend in the preset comparison database that is the same as the short-term behavior trend, the data group to be processed with the earliest storage time in the first unit is directly stored as a historical data group.

[0120] This application also provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed by any of the above-described trend correction methods based on sensor data from a loading device.

[0121] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A trend-based deviation correction method based on loading equipment sensor data, characterized by, The method comprises the following steps: regularly obtaining original data, and generating corresponding to-be-processed data sets from the original data when the number of the original data reaches a first preset value; storing the generated to-be-processed data sets in a preset first unit, and arranging the to-be-processed data sets in the first unit in the order of storage; analyzing a current behavior trend of the to-be-processed data sets in the first unit and a recent behavior trend of historical data sets in a preset second unit whenever the number of the to-be-processed data sets in the first unit exceeds a second preset value; combining the current behavior trend and the recent behavior trend into a short-term behavior trend, and judging whether there is a special behavior trend identical to the short-term behavior trend in a preset comparison database, wherein the comparison database stores a plurality of special behavior trends; if it is judged that there is a special behavior trend identical to the short-term behavior trend in the comparison database, supplementing the to-be-processed data set stored earliest in the first unit, and storing the supplemented to-be-processed data set as a historical data set; if it is judged that there is no special behavior trend identical to the short-term behavior trend in the comparison database, directly storing the to-be-processed data set stored earliest in the first unit as a historical data set; the original data comprises position information and weight information, and the generation of the corresponding to-be-processed data sets from the original data comprises the following steps: generating a blank set comprising a plurality of empty first reserved bits and a plurality of empty second reserved bits, wherein each first reserved bit is used for storing a position information, and each second reserved bit is used for storing a weight information; storing the corresponding position information in the first reserved bits in the order of acquisition time of the original data; storing the corresponding weight information in the second reserved bits in the order of acquisition time of the original data; converting the blank set in which the first reserved bits and the second reserved bits both store corresponding position information and weight information into a to-be-processed data set; the supplementing of the to-be-processed data set stored earliest in the first unit comprises the following steps: judging whether there is an empty value in the first reserved bits and whether there is an empty value in the second reserved bits, if there is an empty value in the first reserved bits, generating a position estimated value according to a preset first supplement rule to replace the empty value in the first reserved bits; if there is an empty value in the second reserved bits, generating a weight estimated value according to a preset second supplement rule to replace the empty value in the second reserved bits; if there are no empty values in the first reserved bits and the second reserved bits, storing the to-be-processed data set as a historical data set; the second supplement rule comprises: defining consecutive empty values as empty sets, taking the weight information at both ends of each empty set as weight reference information of the corresponding empty set in the order of storage time, and replacing the empty values in the empty sets with the two corresponding weight reference information, wherein each empty value is replaced with the weight reference information stored at a similar time in priority.

2. The trend-based deviation correction method based on loading equipment sensor data according to claim 1, wherein, before the corresponding position information is stored in the first reserved bits in the order of acquisition time of the original data, the method further comprises the following steps: judging whether the position information is within a preset first range, If the position information is within the preset first range, the position information is reserved; If the position information is not within the preset first range, the corresponding position information is replaced with a null value; Before the corresponding weight information is sequentially stored into the plurality of second reserved bits according to the order of the original data acquisition time, the method further comprises the following steps: determining whether the weight information is within a preset second range, If the weight information is within the preset second range, the weight information is reserved; If the weight information is not within the preset second range, the corresponding weight information is replaced with a null value.

3. The trend-based deviation correction method based on loading equipment sensor data according to claim 1, wherein, The first supplement rule comprises: defining consecutive null values as an empty set, taking the position information at both ends of each empty set as the position reference information of the corresponding empty set according to the stored time sequence, generating position estimation information according to the two corresponding position reference information, and replacing the null values in the corresponding empty set with the position estimation information.

4. The trend-based deviation correction method of claim 3, wherein, The generating of the position estimation information according to the two corresponding position reference information comprises the following steps: determining whether the distance value of the two corresponding position reference information exceeds a third preset value, If the distance value of the two corresponding position reference information exceeds the third preset value, generating the same number of position estimation information as the number of null values in the corresponding empty set according to the two corresponding position reference information, and sequentially replacing the null values in the corresponding empty set with the position estimation information; If the distance value of the two corresponding position reference information is less than or equal to the third preset value, calculating the average value of the two corresponding position reference information as the position estimation information, and replacing all null values in the corresponding empty set with the position estimation information.

5. A trend-based deviation correction system for loading equipment based on sensor data, characterized in that, It comprises: a data acquisition module for periodically acquiring original data and generating corresponding to-be-processed data groups according to the original data when the number of original data reaches a first preset value; a data storage module for storing the generated to-be-processed data groups into a preset first unit and arranging the to-be-processed data groups in the first unit according to the storage order; a data analysis module for analyzing the current behavior trend of the to-be-processed data groups in the first unit and the recent behavior trend of the historical data groups in a preset second unit whenever the number of to-be-processed data groups in the first unit exceeds a second preset value; a data comparison module for combining the current behavior trend and the recent behavior trend into a short-term behavior trend and determining whether there is a special behavior trend identical to the short-term behavior trend in a preset comparison database, wherein the comparison database stores a plurality of special behavior trends; If it is determined that there is a special behavior trend identical to the short-term behavior trend in the preset comparison database, the to-be-processed data group stored earliest in the first unit is supplemented, and the supplemented to-be-processed data group is stored as a historical data group; If it is determined that there is no special behavior trend identical to the short-term behavior trend in the preset comparison database, the to-be-processed data group stored earliest in the first unit is directly stored as a historical data group; The original data comprises position information and weight information, and the generation of the corresponding to-be-processed data groups according to the original data comprises the following steps: Generate a blank set, which includes a plurality of empty first reserved bits and empty second reserved bits, wherein each first reserved bit is used to store a position information, and each second reserved bit is used to store a weight information; Store the corresponding position information into the plurality of first reserved bits in turn according to the order of the original data acquisition time; Store the corresponding weight information into the plurality of second reserved bits in turn according to the order of the original data acquisition time; Convert the blank set in which the first reserved bits and the second reserved bits both store the corresponding position information and weight information into a to-be-processed data group; The first unit stores the earliest to-be-processed data group in time, which includes the following steps: Determine whether there is a null value in the first reserved bit and whether there is a null value in the second reserved bit, If there is a null value in the first reserved bit, generate a position estimated value according to a preset first supplement rule to replace the null value in the first reserved bit; If there is a null value in the second reserved bit, generate a weight estimated value according to a preset second supplement rule to replace the null value in the second reserved bit; If there is no null value in the first reserved bit and the second reserved bit, store the to-be-processed data group as a historical data group; The second supplement rule includes: defining consecutive null values as a null set, taking the weight information at both ends of each null set as the weight reference information of the corresponding null set according to the storage time order, and replacing the null values in the null set using the two corresponding weight reference information, wherein each null value is replaced by the weight reference information stored at a similar time.

6. A computer-readable storage medium, characterized in that, A computer program is stored, which can be loaded and executed by a processor to implement a trend correction method based on loading equipment sensing data according to any one of claims 1 to 4.

Images