CN111932031B - Cargo state prediction method and device and multi-classification modeling method - Google Patents
Cargo state prediction method and device and multi-classification modeling method Download PDFInfo
- Publication number
- CN111932031B CN111932031B CN202010956370.8A CN202010956370A CN111932031B CN 111932031 B CN111932031 B CN 111932031B CN 202010956370 A CN202010956370 A CN 202010956370A CN 111932031 B CN111932031 B CN 111932031B
- Authority
- CN
- China
- Prior art keywords
- data
- segment
- original
- state
- cargo
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a cargo state prediction method, a cargo state prediction device and a multi-classification modeling method, wherein the cargo state prediction method comprises the steps of obtaining N continuous original data points which are generated by a vehicle-mounted weighing device at the latest, and recording original measurement data in the original data points; obtaining 1 whole segment of data and M segment data according to the continuous N original data points; and inputting the 1 whole segment of data and the M segmented data into a multi-classification model to obtain the current prediction state, wherein the multi-classification model is a model capable of predicting the current state of the goods according to the original measurement data of the vehicle-mounted weighing equipment. The method and the device are used for solving the problems that the timeliness and the accuracy of the existing goods state identification method based on manual identification are poor and the cost is high.
Description
Technical Field
The application relates to the technical field of logistics, in particular to a goods state prediction method and device and a multi-classification modeling method.
Background
In the logistics field, especially in the truck freight field or the trunk road freight field, the judgment on the freight state is very important, and the accurate and timely recognition on the freight state can help a manager to improve the vehicle management efficiency and reduce the transportation cost.
The current cargo state identification is mainly realized through manual identification, namely, a driver or a station manager reports the actual cargo state after manually counting, but the mode has information hysteresis; if the station administrator is required to manually report the cargo state in real time, the labor cost is greatly increased; if the driver is required to report the cargo state in real time, the effect is unsatisfactory, because the driver drives the vehicle at the job, the driver needs to rest after the vehicle is driven to the platform, and the driver does not pay attention to the cargo state, and if the driver is required to report the cargo state in real time, the rest of the driver is inevitably influenced, so that potential safety hazards are caused.
In summary, the existing mode of identifying the state of the goods based on manual identification has poor timeliness and accuracy, and is high in cost.
Disclosure of Invention
The application mainly aims to provide a cargo state prediction method, a cargo state prediction device and a multi-classification modeling method, so as to solve the problems of poor timeliness and accuracy and high cost of the conventional cargo state recognition method based on manual recognition.
In order to achieve the above object, according to a first aspect of the present application, a method of cargo state prediction is provided.
The cargo state prediction method according to the application comprises the following steps:
acquiring N continuous original data points newly generated by vehicle-mounted weighing equipment, wherein original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
obtaining 1 whole segment data and M segment data according to continuous N original data points, wherein M is an integer greater than or equal to 2;
and inputting the 1 whole segment of data and the M segmented data into a multi-classification model to obtain the current prediction state, wherein the multi-classification model is a model capable of predicting the current state of the goods according to the original measurement data of the vehicle-mounted weighing equipment.
Optionally, the inputting 1 whole segment of data and M segment data into the multi-classification model further includes:
and respectively processing the 1 whole segment of data and the M sections of data to obtain processed data, and inputting the processed data into the multi-classification model.
Optionally, the processing the 1 whole segment of data and the M segment of data to obtain processed data, and inputting the processed data into the multi-classification model, further includes:
and respectively carrying out characterization processing on the 1 whole segment data and the M segmented data to obtain characteristic data, and inputting the characteristic data into the multi-classification model.
Optionally, the raw measurement data includes at least one of deformation data, resistance data, voltage data, current data, or time data.
To achieve the above object, according to a second aspect of the present application, there is provided a method of multi-class modeling.
The method for multi-classification modeling according to the application comprises the following steps:
acquiring N continuous original data points and corresponding cargo states from a data set for multiple times, wherein the data set comprises a plurality of original data points and corresponding cargo states, original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
respectively obtaining 1 whole segment of data, M segmented data and state data according to each continuous N original data points and corresponding goods states, wherein M is an integer greater than or equal to 2;
and inputting all 1 whole segment data, M segment data and state data serving as training data into the multi-classification model to be trained to obtain the trained multi-classification model.
Optionally, the obtaining M pieces of segmented data according to each consecutive N original data points includes:
and carrying out segmentation processing on the N original data points according to the way that each segmented data and any other segmented data have data point intersection to obtain M segmented data.
Optionally, obtaining the state data according to each consecutive N original data points and the corresponding cargo states respectively includes:
taking the cargo state corresponding to the last original data point in the N original data points as state data; or the like, or, alternatively,
and determining state data according to the number or the proportion of the corresponding cargo states in the N original data points.
Optionally, inputting all 1 whole segment data, M segment data and state data as training data into the multi-class model to be trained, further comprising:
and performing characterization processing on each of the 1 whole segment data and the M segment data to obtain feature data, and inputting the feature data and the state data corresponding to each of the 1 whole segment data and the M segment data to the multi-classification model to be trained.
In order to achieve the above object, according to a third aspect of the present application, there is provided an apparatus for cargo state prediction.
The cargo state prediction device according to the application comprises:
the system comprises an acquisition unit, a data acquisition unit and a data processing unit, wherein the acquisition unit is used for acquiring N continuous original data points newly generated by the vehicle-mounted weighing equipment, original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
the data processing unit is used for processing the data acquired by the acquisition unit to obtain input data required by the multi-classification model;
the prediction unit is used for inputting the input data obtained by the data processing unit into a multi-classification model to obtain a current prediction state, wherein the multi-classification model is a model capable of predicting the current state of the cargo according to original measurement data of the vehicle-mounted weighing equipment or characteristic data derived from the original measurement data; the prediction unit stores a trained multi-classification model.
Optionally, the data processing unit includes:
the segmentation module is used for obtaining 1 whole segment of data and M segments of data according to the N continuous original data points obtained by the obtaining unit, wherein M is an integer greater than or equal to 2;
and the characteristic module is used for respectively carrying out characterization processing on the 1 whole segment of data and/or the M segments of data obtained by the segmentation module to obtain characteristic data.
In order to achieve the above object, according to a fourth aspect of the present application, there is provided an apparatus for multi-class modeling, characterized in that the apparatus comprises:
the system comprises an acquisition unit, a data acquisition unit and a control unit, wherein the acquisition unit is used for acquiring N continuous original data points and corresponding cargo states from a data set for multiple times, the data set comprises a plurality of original data points and corresponding cargo states, original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
the data processing unit is used for respectively obtaining 1 whole segment of data, M sections of data and state data according to each continuous N original data points and corresponding goods states, wherein M is an integer greater than or equal to 2;
and the training unit is used for inputting all 1 whole segment of data, M pieces of segment data and state data serving as training data into the multi-classification model to be trained to obtain the trained multi-classification model.
In order to achieve the above object, according to a fifth aspect of the present application, there is provided a cargo state prediction method, the method including:
acquiring N continuous original data points newly generated by vehicle-mounted weighing equipment, wherein original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
carrying out characterization processing on the continuous N original data points to obtain characteristic data;
and inputting the characteristic data into a multi-classification model to predict the cargo state.
In order to achieve the above object, according to a sixth aspect of the present application, there is provided a computer-readable storage medium storing computer instructions for causing the computer to execute the cargo state prediction method according to any one of the first or fifth aspects and/or the multi-classification modeling method according to any one of the second aspects.
In order to achieve the above object, according to a seventh aspect of the present application, there is provided an electronic apparatus comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the method of cargo state prediction of any of the first or fifth aspects above and/or the method of multi-class modeling of any of the second aspects.
According to the method and the device for predicting the cargo state, N continuous original data points which are generated by a vehicle-mounted weighing device at the latest are obtained, and original measurement data are recorded in the original data points; obtaining 1 whole segment of data and M segment data according to the continuous N original data points; and inputting 1 whole segment of data and M segmented data into the multi-classification model to obtain the current prediction state. The multi-classification model is a model capable of predicting the current state of the cargo according to the original measurement data of the vehicle-mounted weighing equipment.
In the multi-classification modeling method, N continuous original data points and corresponding cargo states are obtained from a data set for multiple times, the data set comprises a plurality of original data points and corresponding cargo states, and original measurement data are recorded in the original data points; respectively obtaining 1 whole segment of data, M segmented data and state data according to each continuous N original data points and corresponding goods states; and inputting all 1 whole segment data, M segment data and state data serving as training data into the multi-classification model to be trained to obtain the trained multi-classification model.
When the goods state is predicted, the goods state is predicted according to the multi-classification model, manual identification is not needed in the prediction process, the accuracy and timeliness of goods state identification can be obviously improved, and meanwhile the cost is reduced. In addition, the goods state can be predicted only by using the original data of the vehicle-mounted weighing device during prediction, other series of data such as vehicle speed and GPS do not need to be acquired, and a separate prediction model does not need to be established for a specific vehicle.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, serve to provide a further understanding of the application and to enable other features, objects, and advantages of the application to be more apparent. The drawings and their description illustrate the embodiments of the invention and do not limit it. In the drawings:
FIG. 1 is a flow chart of a method of multi-class modeling provided in accordance with an embodiment of the present application;
FIG. 2 is a flow chart of a method for cargo state prediction according to an embodiment of the present application;
fig. 3 is a block diagram of a cargo state prediction apparatus according to an embodiment of the present application;
FIG. 4 is a block diagram of another cargo state prediction apparatus provided in accordance with an embodiment of the present application;
FIG. 5 is a block diagram illustrating components of an apparatus for multi-class modeling according to an embodiment of the present application.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances such that embodiments of the application described herein may be used. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
According to an embodiment of the present application, there is provided a method for multi-class modeling, as shown in fig. 1, the method including the steps of:
s101, acquiring continuous N original data points and corresponding cargo states from a data set for multiple times.
The cargo state corresponding to each original data point in the data set is obtained by data labeling, for example, a plurality of original data points generated for a vehicle-mounted weighing device in the historical data, and the cargo states are various (which can be set manually), during the data labeling, the cargo state to be predicted can be labeled according to needs, if the cargo state to be predicted is loading start, loading finish, unloading start and unloading finish, the four cargo states of loading start, loading finish, unloading start and unloading finish can be labeled on the original data points in the historical data according to actual conditions, and other original data points which are not matched with the four cargo states in the actual conditions can be labeled, so that the data set is formed.
For example, a certain raw data point may correspond to a cargo state, such as the cargo state corresponding to raw data point 1 is another state, the cargo state corresponding to raw data point 2 is another state, the cargo state corresponding to raw data point 3 is a loading start state, and so on. The data set includes a plurality of raw data points and corresponding cargo states, and in this example, the raw data points and corresponding cargo states include five types, namely, loading, unloading, and unloading. It should be noted that the above five cargo states are an example, and in practical applications, the cargo states may also be adaptively adjusted according to actual requirements.
Further, in order to ensure the accuracy of the model, the number of the original data points corresponding to each cargo state is set in the data set. Taking the above five cargo states as an example, the number of the raw data points corresponding to each cargo state is not less than 10% of the total number of the raw data points in the data set (specifically, the number of the raw data points in the data set can be adjusted by a random down-sampling manner); or the number of the original data points corresponding to the four states of loading, loading completion, unloading start and unloading completion is not less than 10% of the total number of the original data points in the data set; or the ratio of the original data points corresponding to the five states of starting loading, completing loading, starting unloading, completing unloading and the like is 1:1:1:1: 2. In the above example, the specific data values of 10% and 1:1:1:1:2 may be adaptively adjusted according to the type of the cargo state and the actual requirement in the actual application.
The raw measurement data generated by the vehicle-mounted weighing device is recorded in the raw data points. For example, a vehicle-mounted weighing apparatus that obtains weighing data (or weight data) through a specific calculation process using deformation data of an object first obtains deformation data of the object through its own measuring device, and then obtains weighing data by calculation using the deformation data, where the original measurement data is the deformation data of the object; for another example, the vehicle-mounted weighing apparatus that obtains the weighing data through a specific calculation process using the resistance data of the object first obtains the resistance data of the object through its own measuring device, and then calculates the weighing data using the resistance data, where the original measuring data is the resistance data of the object; of course, since there are also vehicle-mounted weighing devices manufactured by using other measurement principles, such as those manufactured by using data such as measured current or voltage, and those manufactured by using time data (such as ultrasonic round trip time) generated when a distance is measured by ultrasonic waves, it is only necessary to use raw measurement data in the vehicle-mounted weighing devices.
It should be noted that the original measurement data is not the weighing data or the weight data finally obtained by the vehicle-mounted weighing device, because the existing vehicle-mounted weighing device can adjust the calculation process by combining the influence factors such as the ambient temperature and the vehicle speed, and further the weighing data is close to the real value of the weight of the goods in real time, the same weighing data can be generated by calculating different original measurement data, and if the weighing data is used, a prediction model cannot be well established.
Since the raw data points and their corresponding cargo states in the data set are derived from historical data, the order relationship between each raw data point is also known (e.g., the order relationship derived from the time the raw data points were generated), and thus, N consecutive raw data points and their corresponding cargo states can also be obtained from the data set. Where N is an integer of 2 or more, for example, N =30, 40, or the like can be used.
One continuous N original data points and corresponding cargo states thereof cannot meet the requirement of data quantity of model training, and the more training data, the higher the accuracy of the finally obtained model, so that a plurality of continuous N original data points and corresponding cargo states thereof are required to obtain the training data.
And S102, respectively obtaining 1 whole segment of data, M segmented data and state data according to each continuous N original data points and corresponding goods states.
Wherein 1 entire segment of data contains N consecutive original data points. The M (M is an integer greater than or equal to 2) pieces of segment data are obtained by carrying out segment processing on N original data points according to the way that each piece of segment data and any other piece of segment data have data point intersection. That is, each piece of data includes consecutive ones of the N consecutive original data points, and each piece of data intersects at least one other piece of segmented data. It should be noted that segmenting the data is also performed to improve the accuracy of the model.
Specific examples are given for illustration, for example, N =30, M =3, and the 3 segments of segmented data are: segment data 1:1 st to 16 th raw data points; segment data 2: 7 th to 23 rd raw data points; segment data 3: 15 th to 30 th original data point.
In this embodiment, two determination modes of state data are given:
firstly, a cargo state corresponding to the last original data point in the continuous N original data points is used as state data (if the cargo state corresponding to the last original data point is loading start, the state data is loading start);
second, the number or the ratio of the cargo states corresponding to the N raw data points may be determined (for example, 10 cargo states corresponding to the raw data points in the 30 raw data points are other states, and the cargo state corresponding to the 20 raw data points is the loading starting state, and since the number of raw data points corresponding to the loading starting state is larger, the state data is the loading starting state, that is, the cargo state with the largest number of cargo states corresponding to the N raw data points is used as the state data).
In addition, during the process of obtaining the whole segment of data, the segment data and the state data, the original data point and the corresponding cargo state thereof may be preprocessed according to actual needs, such as format adjustment, digitization and the like, so as to form the whole segment of data, the segment data and the state data required by the model training. The pretreatment method can be realized by utilizing the prior art.
And S103, inputting all 1 whole segment of data, M pieces of segment data and state data serving as training data into the multi-classification model to be trained to obtain the trained multi-classification model.
And inputting all the 1 whole segment data, the M segment data and the state data obtained in the previous step as training data into a multi-classification model to be trained for training to obtain the multi-classification model. The input of the multi-classification model is 1 whole section of data and M sections of data, and the output result of the model is the current cargo state. For the foregoing example, the output of the trained model is one of loading, loading completion, unloading start, unloading completion, and others (other states may be unnecessary for prediction, etc.). The specific model training methods are many, and are not limited herein, as long as the trained multi-class model can be obtained. In addition, the multi-classification model to be trained can be any model capable of realizing multi-classification.
In addition, the 1 whole segment of data and the M pieces of segment data can be processed respectively to obtain processed data, and the processed data and the state data are input into a multi-classification model to be trained to obtain the trained multi-classification model. More specifically, the processing may include preprocessing and/or characterization, or other processing.
Further, before all 1 whole segment data, M segment data, and state data are input to the multi-classification model to be trained as training data, each of the 1 whole segment data and M segment data in all 1 whole segment data and M segment data may be subjected to characterization processing to obtain feature data, and then the feature data and state data corresponding to each of the 1 whole segment data and M segment data are input to the multi-classification model to be trained, so as to obtain the multi-classification model. Specifically, the characterization process is explained as follows:
(1) and taking the median of each segment of data as characteristic data.
Namely, the median of 1 whole segment data and M segment data is used as the characteristic data. With reference to the above specific example, 1 whole segment of data is 30 raw measurement data, and 3 segment data are segment data 1:1 st to 16 th raw measurement data; segment data 2: 7 th to 23 rd raw measurement data; segment data 3: 15 th to 30 th raw measurement data. And obtaining the median of each segment of data according to a median calculation mode.
(2) The ratio of the median of each piece of data was taken as the feature data.
Obtaining the median of each segment of data according to the mode in the step (1), and then calculating the median of each segment pairwise to obtain the ratio of the medias to obtain the characteristic data.
(3) And taking the standard deviation of the difference value of each piece of data as characteristic data.
And calculating the difference value of each section of data to obtain a plurality of difference values, and calculating the standard deviation of the plurality of difference values to obtain the characteristic data.
In addition, the above-mentioned (1) to (3) are three modes of the characterization processing, and in practical application, any one or a combination of a plurality of modes may be selected for the characterization processing.
From the above description, it can be seen that in the multi-class modeling method according to the embodiment of the present application, when data processing is performed, the selected raw data point is the raw measurement value of the vehicle-mounted weighing device, and segmentation processing is also performed, so that the accuracy of the model can be improved.
In addition, it should be noted that, in practical applications, data related to the cargo state, such as vehicle speed data and vehicle positioning data, may also be added as input data of the model during training in the multi-class modeling, which still belongs to the protection scope of the present technical solution.
In addition, the embodiment of the application also provides another multi-classification modeling method, which comprises the following steps:
firstly, acquiring N continuous original data points and corresponding goods states from a data set for multiple times, wherein the data set comprises a plurality of original data points and corresponding goods states, original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
the implementation manner of this step can be referred to the implementation manner of step S101 in fig. 1, and is not described here again.
Secondly, respectively obtaining 1 whole section of data and state data according to each continuous N original data points and corresponding goods states;
the implementation of this step is different from the implementation of step S102 in fig. 1 in that it is not necessary to segment one whole segment of data to obtain M pieces of segmented data, and the implementation of obtaining 1 whole segment of data and state data is the same, and is not described herein again. It should be noted that, the multi-classification modeling method is different from the aforementioned multi-classification modeling method in that M pieces of segment data are not obtained here, and M pieces of segment data are not used in the subsequent training, and other processing methods are similar to each other.
And finally, performing characterization processing on each 1 whole segment of data to obtain feature data, and inputting the feature data and the state data corresponding to each 1 whole segment of data into the multi-classification model to be trained to obtain the multi-classification model.
The implementation manner of performing the characterization processing on each 1 whole segment of data to obtain the feature data is the same as the implementation manner of performing the characterization processing on 1 whole segment of data and M segment data in the foregoing embodiment, and details are not described here. And after the characteristic data are obtained, inputting the characteristic data and the state data corresponding to all 1 whole segment of data into a multi-classification model to be trained for training to obtain the multi-classification model. The input of the multi-classification model is 1 whole data, and the output result of the model is the current cargo state. For the foregoing example, the output of the trained model is one of loading, loading completion, unloading start, unloading completion, and others (other states may be unnecessary for prediction, etc.). The specific model training methods are many, and are not limited herein, as long as the trained multi-class model can be obtained. In addition, the multi-classification model to be trained can be any model capable of realizing multi-classification.
According to an embodiment of the present application, there is provided a cargo state prediction method, as shown in fig. 2, the method includes the following steps:
s201, acquiring N continuous original data points newly generated by the vehicle-mounted weighing equipment, and recording original measurement data in the original data points.
It should be noted that, the value of N needs to be consistent with the value of N selected during the multi-classification model training. The nature of the raw data points is the same as the raw data points involved in the multi-classification model training process. Namely, if the original data points used in the multi-classification model training are deformation data, the deformation data are also acquired in the step; if the resistance data is obtained, the resistance data is also obtained; and the like.
S202, obtaining 1 whole segment of data and M segments of data according to the continuous N original data points.
The implementation manner of this step is the same as the manner of obtaining 1 whole segment of data and M segment of data according to the consecutive N original data points in step S102 in fig. 1, and is not described herein again. And the values of N and M are the same as those of N and M selected in actual training.
S203, inputting 1 whole segment of data and M segment data into a multi-classification model to obtain the current prediction state.
The 1 whole segment data and the M segment data obtained in step S202 are input into the multi-classification model trained in the embodiment of fig. 1 to be predicted, so as to obtain a current predicted state, that is, a current cargo state. And outputting any one of the results when the current prediction state is the multi-classification model training. For example, the current predicted status may be one of start of loading, completion of loading, start of unloading, completion of unloading, and others.
In addition, the 1 whole segment data and the M segment data may be processed to obtain processed data, and the processed data may be input to the multi-classification model. More specifically, the processing may include preprocessing and/or characterization, or other processing.
In addition, corresponding to the training process of the multi-classification model, if the process of preprocessing and/or characterizing 1 whole data and M segmented data is performed, the 1 whole data and M segmented data obtained from the N continuous original data points that are generated most recently need to be preprocessed and/or characterized in the same manner during the prediction. Namely, 1 whole segment of data and M segment data are respectively subjected to characterization processing to obtain feature data, and the feature data are input into a trained multi-classification model to obtain a current prediction state.
From the above description, it can be seen that, in the method for predicting the cargo state of the present embodiment, N consecutive original data points that are newly generated by the vehicle-mounted weighing device are obtained, and original measurement data is recorded in the original data points; obtaining 1 whole segment of data and M segment data according to the continuous N original data points; and inputting 1 whole segment of data and M segmented data into the multi-classification model to obtain the current prediction state. The multi-classification model is a model capable of predicting the current state of the cargo according to the original measurement data of the vehicle-mounted weighing equipment. Specifically, in the multi-classification modeling method, continuous N original data points and corresponding cargo states thereof are obtained from a data set for multiple times, the data set comprises a plurality of original data points and corresponding cargo states thereof, and original measurement data are recorded in the original data points; respectively obtaining 1 whole segment of data, M segmented data and state data according to each continuous N original data points and corresponding goods states; and inputting all 1 whole segment data, M segment data and state data serving as training data into the multi-classification model to be trained to obtain the trained multi-classification model.
Therefore, when the goods state is predicted in the application, the goods state is predicted according to the multi-classification model, and manual identification is not needed in the prediction process, so that the accuracy and timeliness of goods state identification can be obviously improved, and meanwhile, the cost can be reduced. In addition, the goods state can be predicted only by using the original data of the vehicle-mounted weighing equipment during prediction, other series of data such as vehicle speed and GPS do not need to be acquired, and a separate prediction model does not need to be established for a specific vehicle.
Besides, the embodiment of the application also provides a cargo state prediction method, which specifically comprises the following steps.
Firstly, acquiring N continuous original data points newly generated by vehicle-mounted weighing equipment, wherein original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2;
the implementation of this step is the same as that of step S201 in fig. 2, and is not described here again.
And secondly, performing characterization processing on the continuous N original data points to obtain feature data, and inputting the feature data into a multi-classification model to obtain the current prediction state.
It should be noted that, performing characterization processing on the N consecutive original data points to obtain feature data is equivalent to obtaining 1 whole data according to the N consecutive original data points, and performing characterization processing on 1 whole data to obtain feature data. Specifically, the processing mode of performing the characterization processing on 1 entire segment of data to obtain the feature data is the same as the mode of performing the characterization processing on 1 entire segment of data during the multi-classification model training, which is not described herein again, and the corresponding feature data is obtained after the characterization processing. And then inputting the obtained characteristic data into a multi-classification model obtained by training before to obtain a current prediction state, namely a current cargo state. And outputting any one of the results when the current prediction state is the multi-classification model training. For example, the current predicted status may be one of start of loading, completion of loading, start of unloading, completion of unloading, and others. It should be noted that the multi-classification model to which feature data is input here is a multi-classification model obtained by inputting feature data and state data corresponding to 1 entire piece of data into the multi-classification model to be trained.
From the above description, it can be seen that, in the method for predicting the cargo state according to the embodiment of the present application, N consecutive original data points newly generated by a vehicle-mounted weighing device are obtained, and original measurement data is recorded in the original data points; and performing characterization processing on the continuous N original data points to obtain characteristic data, and inputting the characteristic data into the multi-classification model to obtain the current prediction state. The multi-classification model is a model capable of predicting the current state of the cargo according to the original measurement data of the vehicle-mounted weighing equipment. Specifically, in the multi-classification modeling method, continuous N original data points and corresponding cargo states thereof are obtained from a data set for multiple times, the data set comprises a plurality of original data points and corresponding cargo states thereof, and original measurement data are recorded in the original data points; respectively obtaining 1 whole section of data and state data according to each continuous N original data points and the corresponding cargo state; and performing characterization processing on each 1 whole segment of data to obtain feature data, and inputting the feature data and the state data corresponding to each 1 whole segment of data into the multi-classification model to be trained to obtain the trained multi-classification model.
Therefore, when the goods state is predicted in the application, the goods state is predicted according to the multi-classification model, and manual identification is not needed in the prediction process, so that the accuracy and timeliness of goods state identification can be improved, and meanwhile, the cost can be reduced. In addition, the goods state can be predicted only by using the original data of the vehicle-mounted weighing equipment during prediction, other series of data such as vehicle speed and GPS do not need to be acquired, and a separate prediction model does not need to be established for a specific vehicle.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.
According to an embodiment of the present application, there is also provided an apparatus for cargo state prediction for implementing the method of fig. 2, as shown in fig. 3, the apparatus includes:
the acquiring unit 31 is configured to acquire N consecutive original data points that are newly generated by the vehicle-mounted weighing device, where the original data points record original measurement data, and N is an integer greater than or equal to 2;
the data processing unit 32 is configured to process the data acquired by the acquisition unit 31 to obtain input data required by the multi-classification model;
a prediction unit 33, configured to input the input data obtained by the data processing unit 32 into the multi-classification model to obtain a current prediction state;
the multi-classification model is a model which can predict the current state of the cargo according to original measurement data of the vehicle-mounted weighing equipment or characteristic data derived from the original measurement data;
the prediction unit stores a trained multi-classification model.
Further, as shown in fig. 4, the data processing unit 32 further includes,
a segmenting module 321, configured to obtain 1 whole segment of data and M segments of data according to N consecutive original data points obtained by the obtaining unit 31, where M is an integer greater than or equal to 2;
further, as shown in fig. 4, the data processing unit 32 further includes,
the feature module 322 is configured to perform a characterization process on the 1 whole segment of data and/or the M segments of data obtained by the segmentation module 321, respectively, to obtain feature data.
Specifically, the specific process of implementing the functions of each unit and module in the device in the embodiment of the present application may refer to the related description in the method embodiment, and is not described herein again.
There is also provided, according to an embodiment of the present application, an apparatus for multi-class modeling for implementing the method of fig. 1, as shown in fig. 5, the apparatus including:
an obtaining unit 41, configured to obtain N consecutive original data points and corresponding cargo states thereof from a data set multiple times, where the data set includes the original data points and the corresponding cargo states thereof, the original data points record original measurement data, and N is an integer greater than or equal to 2;
the data processing unit 42 is configured to obtain 1 whole segment of data, M segment of data, and state data according to each consecutive N original data points and the corresponding cargo state, where M is an integer greater than or equal to 2;
and the training unit 43 is configured to input all 1 whole segment of data, M segment data, and state data as training data into the multi-class model to be trained, so as to obtain the trained multi-class model.
Specifically, the specific process of implementing the functions of each unit and module in the device in the embodiment of the present application may refer to the related description in the method embodiment, and is not described herein again.
From the above description, it can be seen that, in the multi-class modeling apparatus according to the embodiment of the present application, when data processing is performed, the selected raw data point is the raw measurement value of the vehicle-mounted weighing device, and segmentation processing is also performed, so that the accuracy of the model can be improved.
According to an embodiment of the present application, there is further provided a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions for causing the computer to execute the method for cargo state prediction and/or the method for multi-class modeling in the above method embodiment.
According to an embodiment of the present application, there is also provided an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the method of cargo state prediction and/or the method of multi-classification modeling in the above method embodiments.
It will be apparent to those skilled in the art that the modules or steps of the present application described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and they may alternatively be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, or fabricated separately as individual integrated circuit modules, or fabricated as a single integrated circuit module from multiple modules or steps. Thus, the present application is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (5)
1. A method of cargo state prediction, the method comprising:
acquiring N continuous original data points newly generated by the vehicle-mounted weighing equipment, wherein original measurement data generated by the vehicle-mounted weighing equipment are recorded in the original data points, and N is an integer greater than or equal to 2;
obtaining 1 whole segment data and M segment data according to continuous N original data points, wherein M is an integer greater than or equal to 2; the M pieces of segment data are obtained by carrying out segment processing on N original data points in a way that each piece of segment data and any other piece of segment data have data point intersection;
respectively performing characterization processing on 1 whole segment of data and M pieces of segment data to obtain feature data, and inputting the feature data into a multi-classification model to obtain a current prediction state, wherein the multi-classification model is a model capable of predicting the current state of goods according to original measurement data of vehicle-mounted weighing equipment;
the modeling method of the multi-classification model comprises the following steps: acquiring N continuous original data points and corresponding cargo states from a data set for multiple times, wherein the data set comprises a plurality of original data points and corresponding cargo states, the number of the original data points corresponding to each cargo state is not less than 10% of the total number of the original data points in the data set, original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2; the step of obtaining N continuous original data points and corresponding cargo states from the data set for multiple times comprises the step of taking the cargo state corresponding to the last original data point in the N original data points as state data; or determining state data according to the number or proportion of the corresponding cargo states in the N original data points; respectively obtaining 1 whole segment of data, M segmented data and state data according to each continuous N original data points and corresponding goods states, wherein M is an integer greater than or equal to 2; the M pieces of segment data are obtained by carrying out segment processing on N original data points in a way that each piece of segment data and any other piece of segment data have data point intersection;
and performing characterization processing on each of the 1 whole segment data and the M segment data to obtain feature data, and inputting the feature data and the state data corresponding to each of the 1 whole segment data and the M segment data to a multi-classification model to be trained to obtain the trained multi-classification model.
2. The method of cargo state prediction according to claim 1, characterized in that the raw measurement data comprises at least one of deformation data, resistance data, voltage data, current data or time data.
3. A cargo state prediction apparatus for implementing the cargo state prediction method according to any one of claims 1-2, the apparatus comprising:
the system comprises an acquisition unit, a data acquisition unit and a data processing unit, wherein the acquisition unit is used for acquiring N continuous original data points newly generated by the vehicle-mounted weighing equipment, original measurement data are recorded in the original data points, and N is an integer greater than or equal to 2; the M pieces of segment data are obtained by carrying out segment processing on N original data points in a way that each piece of segment data and any other piece of segment data have data point intersection;
the data processing unit is used for processing the data acquired by the acquisition unit to obtain input data required by the multi-classification model;
the prediction unit is used for inputting the input data obtained by the data processing unit into a multi-classification model to obtain a current prediction state, wherein the multi-classification model is a model capable of predicting the current state of the goods according to original measurement data of the vehicle-mounted weighing equipment or characteristic data derived from the original measurement data, and a trained multi-classification model is stored in the prediction unit;
the segmentation module is used for obtaining 1 whole segment of data and M segments of data according to the N continuous original data points obtained by the obtaining unit, wherein M is an integer greater than or equal to 2; the M pieces of segment data are obtained by carrying out segment processing on N original data points in a way that each piece of segment data and any other piece of segment data have data point intersection;
and the characteristic module is used for respectively carrying out characterization processing on the 1 whole segment of data and/or the M segments of data obtained by the segmentation module to obtain characteristic data.
4. A computer-readable storage medium storing computer instructions for causing a computer to perform the method of cargo state prediction of any of claims 1-2.
5. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the method of cargo state prediction of any of claims 1-2.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010956370.8A CN111932031B (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
| CN202210054273.9A CN114548508A (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010956370.8A CN111932031B (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210054273.9A Division CN114548508A (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111932031A CN111932031A (en) | 2020-11-13 |
| CN111932031B true CN111932031B (en) | 2022-02-08 |
Family
ID=73309913
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210054273.9A Pending CN114548508A (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
| CN202010956370.8A Active CN111932031B (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210054273.9A Pending CN114548508A (en) | 2020-09-11 | 2020-09-11 | Cargo state prediction method and device and multi-classification modeling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN114548508A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609808A (en) * | 2017-07-26 | 2018-01-19 | 深圳市易成自动驾驶技术有限公司 | Monitoring method, device and the computer-readable recording medium of car status information |
| CN107680375A (en) * | 2017-09-29 | 2018-02-09 | 深圳市易成自动驾驶技术有限公司 | A kind of vehicle load computational methods, device and storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070040677A1 (en) * | 2005-08-17 | 2007-02-22 | Blair Herbert W Jr | Detecting cargo status and load activity |
| US20180060774A1 (en) * | 2016-09-01 | 2018-03-01 | Blackberry Limited | Efficiency of a cargo shipping system |
| CN109808839B (en) * | 2017-11-22 | 2020-07-17 | 中冶宝钢技术服务有限公司 | Loading method for ship |
| US11475680B2 (en) * | 2018-12-12 | 2022-10-18 | Spireon, Inc. | Cargo sensor system implemented using neural network |
| CN110991965B (en) * | 2019-11-29 | 2023-04-18 | 广汽丰田汽车有限公司 | Logistics transportation in-transit monitoring management method, system and storage medium |
| CN111209951B (en) * | 2020-01-02 | 2021-03-23 | 深圳市汉德网络科技有限公司 | Real-time vehicle-mounted weighing method |
-
2020
- 2020-09-11 CN CN202210054273.9A patent/CN114548508A/en active Pending
- 2020-09-11 CN CN202010956370.8A patent/CN111932031B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107609808A (en) * | 2017-07-26 | 2018-01-19 | 深圳市易成自动驾驶技术有限公司 | Monitoring method, device and the computer-readable recording medium of car status information |
| CN107680375A (en) * | 2017-09-29 | 2018-02-09 | 深圳市易成自动驾驶技术有限公司 | A kind of vehicle load computational methods, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111932031A (en) | 2020-11-13 |
| CN114548508A (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110288096B (en) | Prediction model training method, prediction model training device, prediction model prediction method, prediction model prediction device, electronic equipment and storage medium | |
| CN104599002B (en) | Method and equipment for predicting order value | |
| CN110689070B (en) | Training method and device of business prediction model | |
| US20190378180A1 (en) | Method and system for generating and using vehicle pricing models | |
| CN112862184A (en) | Transportation information prediction method and device, electronic equipment and readable storage medium | |
| CN112381303A (en) | Task index data prediction method and system | |
| CN111859172A (en) | Information pushing method and device, electronic equipment and computer readable storage medium | |
| CN111932031B (en) | Cargo state prediction method and device and multi-classification modeling method | |
| CN113807560A (en) | Logistics cost prediction method and device, electronic equipment and storage medium | |
| CN111507541B (en) | Goods quantity prediction model construction method, goods quantity measurement device and electronic equipment | |
| CN110598989B (en) | Goods source quality evaluation method, device, equipment and storage medium | |
| CN111932035B (en) | Data processing method and device based on multiple models and classified modeling method | |
| CN112364262A (en) | Goods finding route recommendation method and device, electronic equipment and storage medium | |
| CN116415747A (en) | Method and device for determining carpooling route and electronic equipment | |
| CN111091420A (en) | Method and device for predicting power price | |
| CN111815250A (en) | Goods state identification method and device for logistics and two-classification modeling method | |
| CN111680941A (en) | Premium recommendation method, device, equipment and storage medium | |
| CN115689245A (en) | Method and device for determining rider's saddle bill quantity in different scenes | |
| CN115760187A (en) | Transportation capacity cost estimation method and device, electronic equipment and storage medium | |
| CN112051843B (en) | Path planning method and device based on order prediction, robot and storage medium | |
| CN114925895A (en) | Maintenance equipment prediction method, terminal and storage medium | |
| CN114239996A (en) | Logistics information prediction method, device, equipment, platform and storage medium | |
| CN112308344A (en) | Method and device for predicting reservation value of non-departure flight and electronic equipment | |
| CN111209937A (en) | Charging curve model classification method and device and server | |
| CN113222502A (en) | Railway vehicle position tracking method, device and terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |