CN111341098A - Congestion state prediction method and device - Google Patents
Congestion state prediction method and device Download PDFInfo
- Publication number
- CN111341098A CN111341098A CN202010100665.5A CN202010100665A CN111341098A CN 111341098 A CN111341098 A CN 111341098A CN 202010100665 A CN202010100665 A CN 202010100665A CN 111341098 A CN111341098 A CN 111341098A
- Authority
- CN
- China
- Prior art keywords
- prediction
- congestion state
- data
- sensor
- congestion
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Traffic Control Systems (AREA)
- Devices For Checking Fares Or Tickets At Control Points (AREA)
Abstract
The method comprises the steps of obtaining sensor data before the current moment, wherein the sensor data is data collected by a sensor arranged on each road of a toll station; and performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a preset time, wherein the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree. The application aims to provide a mode for predicting the congestion state of a toll station.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method and an apparatus for predicting a congestion state.
Background
Along with the improvement of national life, the number of automobiles is increased every day, the pressure borne by toll stations is synchronously increased, more than nearly ten thousand toll stations are shared in the whole country, if the congestion state of the toll stations can be predicted in advance, the congestion state of the toll stations can be predicted as early as possible in a peak period, and meanwhile, the number of channels is regulated and controlled through the predicted congestion state in a low-peak period, so that the air channel rate is greatly reduced, and the utilization rate of the toll channels is improved. Therefore, it is highly desirable to provide a way to predict the congestion status of toll booths.
Disclosure of Invention
The application mainly aims to provide a method and a device for predicting a congestion state so as to solve the problem of how to predict the congestion state of a toll station.
In order to achieve the above object, according to a first aspect of the present application, a method of congestion status prediction is provided.
The method for predicting the congestion state comprises the following steps:
acquiring sensor data before the current moment, wherein the sensor data is data acquired by a sensor arranged on each road of a toll station;
and performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a preset time, wherein the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree.
Optionally, the method further includes:
acquiring a target prediction time interval, wherein the target prediction time interval is a time interval which is selected by a user and needs to predict the road congestion state of a toll station, and the target prediction time interval is an integral multiple of a preset time length;
and circularly utilizing the congestion state prediction model to carry out prediction to obtain a congestion state prediction result in a target prediction time period, wherein the number of circulation is equal to the multiple of the integral multiple.
Optionally, the method further includes:
acquiring sensor sample data;
preprocessing the sensor sample data and generating characteristic data corresponding to the sensor sample data;
and training feature data corresponding to the sensor sample data by an integrated machine learning algorithm based on a decision tree to obtain a congestion state prediction model.
Optionally, the preprocessing the sensor sample data and generating the feature data corresponding to the sensor sample data includes:
preprocessing sensor sample data, wherein each piece of data in the sensor sample data obtains corresponding first characteristic data, and the first characteristic data comprises one or more characteristics of a sensor Identity (ID), time characteristic data, current congestion state data and road characteristic data; and the number of the first and second electrodes,
and generating second characteristic data corresponding to each piece of data according to the congestion state before the time corresponding to each piece of data, wherein the second characteristic data comprises one or more characteristics of the congestion state at continuous time points, the accumulated congestion state at continuous time points and the accumulated congestion state at interval time points.
Optionally, the predicting by cyclically using the congestion state prediction model to obtain a prediction result of the congestion state in the target prediction time period includes:
equally dividing the target prediction time interval into a plurality of sub-target prediction time intervals according to a preset time length;
obtaining a first congestion state corresponding to a first sub-target prediction time period by using a congestion state prediction model according to sensor data before the current time;
obtaining a second congestion state corresponding to a second sub-prediction target time period by using the congestion state prediction model again according to the sensor data before the current time, the first sub-prediction target time period and the corresponding first congestion state;
and sequentially acquiring the congestion states corresponding to the residual sub-target prediction time periods by circularly utilizing the congestion state prediction model according to the mode of acquiring the second congestion state.
Optionally, before performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a corresponding congestion state after a predetermined time period, the method further includes:
and preprocessing the sensor data before the current moment and generating characteristic data corresponding to the sensor data before the current moment.
In order to achieve the above object, according to a second aspect of the present application, there is provided an apparatus for congestion state prediction.
The congestion state prediction device according to the application comprises:
the system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring sensor data before the current moment, and the sensor data is data acquired by a sensor arranged on each road of a toll station;
and the prediction unit is used for predicting the congestion state based on a congestion state prediction model according to the sensor data to obtain a prediction result of the congestion state corresponding to the road where the sensor is located after a preset time, and the congestion state prediction model is obtained by training the sensor sample data based on an integrated machine learning algorithm of a decision tree.
Optionally, the apparatus further comprises:
the second acquisition unit is used for acquiring a target prediction time interval, wherein the target prediction time interval is a time interval which is selected by a user and needs to predict the road congestion state of the toll station, and the target prediction time interval is an integral multiple of a preset time length;
the prediction unit is further configured to perform prediction by cyclically using the congestion state prediction model to obtain a congestion state prediction result in a target prediction time period, and the number of times of the cycle is equal to a multiple of the integral multiple.
Optionally, the apparatus further comprises:
a third acquisition unit for acquiring sensor sample data;
the sample characteristic generating unit is used for preprocessing the sensor sample data and generating characteristic data corresponding to the sensor sample data;
and the model generation unit is used for training the feature data corresponding to the sensor sample data based on the integrated machine learning algorithm of the decision tree to obtain a congestion state prediction model.
Optionally, the sample feature generating unit includes:
the system comprises a first characteristic generation module, a second characteristic generation module and a third characteristic generation module, wherein the first characteristic generation module is used for preprocessing sensor sample data, each piece of data in the sensor sample data obtains corresponding first characteristic data, and the first characteristic data comprises one or more characteristics of a sensor Identity (ID), time characteristic data, current congestion state data and road characteristic data;
the second characteristic generating module is used for generating second characteristic data corresponding to each piece of data according to the congestion state before the time corresponding to each piece of data, wherein the second characteristic data comprises one or more characteristics of the congestion state at continuous time points, the accumulated congestion state at continuous time points and the accumulated congestion state at interval time points.
Optionally, the prediction unit includes:
the time interval dividing module is used for equally dividing the target prediction time interval into a plurality of sub-target prediction time intervals according to the preset time length;
the prediction module is used for obtaining a first congestion state corresponding to a first sub-target prediction time period by using a congestion state prediction model according to sensor data before the current moment;
the prediction module is further used for obtaining a second congestion state corresponding to a second sub-prediction target time interval by using the congestion state prediction model again according to the sensor data before the current time, the first sub-prediction target time interval and the corresponding first congestion state;
the prediction module is further configured to sequentially acquire the congestion states corresponding to all the remaining sub-target prediction time periods by cyclically utilizing the congestion state prediction model according to the manner of acquiring the second congestion state.
Optionally, the apparatus further comprises:
and the characteristic determining unit is used for preprocessing the sensor data before the current time and generating the characteristic data corresponding to the sensor data before the current time before the congestion state prediction is carried out based on the congestion state prediction model according to the sensor data and the prediction result of the corresponding congestion state after the preset time length is obtained.
In order to achieve the above object, according to a third aspect of the present application, there is provided a computer-readable storage medium storing computer instructions for causing the computer to execute the method for congestion state prediction according to any one of the first aspects.
In order to achieve the above object, according to a fourth 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 congestion status prediction according to any of the first aspect.
In the method and the device for predicting the congestion state in the embodiment of the application, firstly, sensor data before the current moment is obtained, wherein the sensor data is data acquired by a sensor arranged on each road of a toll station; and performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a preset time, wherein the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree. The method and the device are applied to a road monitoring system of a toll station, can predict the road congestion state in advance according to the sensor data acquired by the installed sensors of all roads, and can adjust the roads and mobilize personnel in time, so that the utilization rate of a toll channel is improved. In addition, when prediction is carried out, prediction is carried out according to a machine algorithm model, and the accuracy and the speed of a prediction result are higher.
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 flowchart of a method for predicting a congestion status according to an embodiment of the present application;
FIG. 2 is a flowchart of a method for generating a congestion status prediction model training according to an embodiment of the present application;
fig. 3 is a block diagram of a congestion status prediction apparatus according to an embodiment of the present application;
fig. 4 is a block diagram of another congestion status prediction apparatus provided 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 predicting a congestion state, as shown in fig. 1, the method including the steps of:
s101, acquiring sensor data before the current moment, wherein the sensor data is acquired by a sensor installed on each road of a toll station.
The sensor system installed on each road of the toll station has a unique identification ID. The sensor data is acquired once every certain time interval, and if the sensor data is acquired once every 5 minutes, the acquired data in one day is 288. Each piece of data includes a sensor ID, a date, a time, a road type, a direction, a congestion level, and the like, where the congestion level may be divided into four types: non, light, medium, heavy (representing unobstructed, slightly congested, moderately congested, heavily congested, respectively). The classification of the congestion level may also be freely adjusted according to the actual demand, and this embodiment is not limited.
And S102, carrying out congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of the congestion state corresponding to the road where the sensor is located after a preset time.
The sensor sample data is the same as the acquired sensor data content before the current time, and includes a sensor ID, date, time, road type, direction, congestion level, and the like. The principle of the congestion state prediction model is to predict the congestion state of a road after a predetermined period of time based on the congestion state of the road before the current time. The congestion state prediction model is a prediction model corresponding to a single sensor. Each road is provided with a sensor, and a prediction model of a single sensor is a congestion state prediction model of the road where the sensor is located.
The congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree. An integrated machine learning algorithm (XGBoost) for a Decision Tree is a general purpose algorithm, and one representative of the algorithm is a Gradient Boosting Decision Tree (GBDT), also known as mart (multiple Additive Regression Tree). The principle of GBDT is that a tree is first trained using a training set and sample truth (i.e. standard answers), and then the training set is predicted using the tree to obtain a predicted value of each sample, and since the predicted value deviates from the truth, the predicted value is subtracted from the truth to obtain a "residual". Next, a second tree is trained, where the truth is no longer used, but the residual is used as the standard answer. After the training of two trees is completed, the residual error of each sample can be obtained again, and then a third tree is further trained, and so on. The total number of trees can be specified manually, or some indicator (e.g., error on the validation set) can be monitored to stop training. When a new sample is predicted, each tree has an output value, and the output values are added to obtain a final predicted value of the sample.
From the above description, it can be seen that, in the method for predicting a congestion state in the embodiment of the present application, first, sensor data before the current time is obtained, where the sensor data is data acquired by a sensor installed on each road of a toll station; and performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a preset time, wherein the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree. The method and the device are applied to a road monitoring system of a toll station, can predict the road congestion state in advance according to the sensor data acquired by the installed sensors of all roads, and can adjust the roads and mobilize personnel in time, so that the utilization rate of a toll channel is improved. In addition, when prediction is carried out, prediction is carried out according to a machine algorithm model, and the accuracy and the speed of a prediction result are higher.
As a further supplement to the above embodiment, the congestion state model in fig. 1 needs to be trained in advance according to sample data, and this embodiment further provides a process for generating a congestion state model by training, as shown in fig. 2, including the following steps:
s201, obtaining sensor sample data.
The sensor sample data is data of a congestion state acquired once at a certain interval, and if the congestion state can be acquired every 5 minutes, one piece of data comprises a sensor ID, a date, time, a road type, a direction, a congestion level and the like, the congestion level is divided into four types, namely unobstructed congestion, slight congestion, moderate congestion and severe congestion, and the specific congestion level can be represented by 0, 1, 2 and 3 respectively.
S202, preprocessing the sensor sample data and generating characteristic data corresponding to the sensor sample data.
The specific implementation of the step comprises the following steps:
preprocessing sensor sample data, wherein each piece of data in the sensor sample data obtains corresponding first characteristic data, and the first characteristic data comprises one or more characteristics of a sensor Identity (ID), time characteristic data, current congestion state data and road characteristic data;
the specific preprocessing comprises the steps of combining date and time into a begain _ time column, sorting according to time, and deleting and correcting abnormal data (a correction method is adopted, namely data of upper and lower time points of a current sensor are adopted to artificially correct the abnormal data); all (sensor IDs) were generated into standardized models by LabelEncoder; LabeleEncoder converts the sensor ID to a serial number. The merged begain _ time column corresponds to time characteristic data, the number behind the LabelEncoder is a sensor identity ID characteristic, the current congestion state data characteristic is (0, 1, 2 and 3), and the road characteristic data comprises the direction of the road and the type of the road.
And generating second characteristic data corresponding to each piece of data according to the congestion state before the time corresponding to each piece of data, wherein the second characteristic data comprises one or more characteristics of the congestion state at continuous time points, the accumulated congestion state at continuous time points and the accumulated congestion state at interval time points. The congestion state of the continuous time points is the congestion state of a plurality of continuous time points before the current time corresponding to each piece of data; the accumulated congestion state of the interval time points is the accumulated congestion state of a plurality of interval time points before the current time corresponding to each piece of data. The second characteristic data is explained below with specific examples:
congestion status characteristics at successive time points: generating ago5, ago4, ago3, go2, ago1 and data characteristics of the first 5 time points through the congestion state and the begain _ time (if one state is recorded in 5 minutes, the data characteristics of the 5 time points are the congestion states of the current 5 minutes, the first 10 minutes, the first 15 minutes, the first 20 minutes and the first 25 minutes of the current data, and if no, the data characteristics are set as Nan); it should be noted that several continuous time points in practical application can be adjusted according to practical requirements, and generally, the more time points are taken, the higher the accuracy of the finally obtained model is.
Cumulative congestion status characteristics at successive time points: corresponding to the above example, the accumulated congestion states at the successive time points are accumulated of the congestion states corresponding to five time points of ago5, ago4, ago3, go2 and ago1, and the accumulated congestion states at the successive time points are characterized by the accumulation of unobstructed-0, light congestion-1, moderate congestion-2 and heavy congestion-3, which can be respectively written as total0, total1, total2 and total 3;
cumulative congestion status characteristics at intervals: congestion states at 5 time points and 10 time points before the current time are accumulated to generate back _ time _5_0, back _ time _5_1, back _ time _5_2 and back _ time _5_ 3;
back_time_10_0，back_time_10_1，back_time_10_2，back_time_10_3；
the accumulated congestion state characteristics of the interval time points are the accumulation of smooth-0, slight congestion-1, moderate congestion-2 and severe congestion-3.
It should be noted that the congestion state accumulation at several time points (5 time points and 10 time points in the example) can be freely adjusted according to actual conditions.
In addition, before preprocessing the sensor sample data, classification is performed according to sensor IDs, and model training is performed on sensor sample data with different IDs, so that it is ensured that the finally obtained congestion state prediction model is a prediction model for a single sensor.
S203, training feature data corresponding to the sensor sample data based on an integrated machine learning algorithm of the decision tree to obtain a congestion state prediction model.
After step S202, generating first feature data and second feature data corresponding to each piece of data, and training the first feature data and the second feature data corresponding to all the sensor sample data based on an integrated machine learning algorithm of a decision tree to obtain a congestion state prediction model. After the training is finished, the congestion state prediction model inputs feature data (first feature data and second feature data) corresponding to the sensor data, and outputs the congestion state after a preset time length.
In addition to the embodiment of fig. 1, it is necessary to add "preprocessing the sensor data before the current time and generating the feature data corresponding to the sensor data before the current time before step 102 is executed in conjunction with the flowchart of fig. 2. And inputting the obtained characteristic data into a congestion state prediction model to obtain the congestion state after a preset time. The process of "preprocessing the sensor data before the current time and generating the feature data corresponding to the sensor data before the current time" is the same as the process of preprocessing the sensor sample data to generate the feature data of the sensor sample data.
In addition, as a supplement to the above embodiment, the present embodiment provides another congestion state prediction method, which is directed to the congestion state within the period that needs to predict the congestion state of the toll gate road selected by the user, i.e., the target prediction period is an integral multiple of the predetermined time length, while the congestion state after the predetermined time length is predicted in fig. 1. To give a specific example for explanation, assuming that the predetermined time period is 5 minutes, the target prediction period may be 10 minutes, 15 minutes, 20 minutes, 25 minutes, 30 minutes, 40 minutes, or the like after the current time. The following describes a process of predicting the congestion state in the target prediction period, taking the target prediction period as 30 minutes after the current time and the predetermined time period as 5 minutes as an example:
firstly, equally dividing a target prediction time interval into a plurality of sub-target prediction time intervals according to a preset time length;
equally dividing 30 minutes into a first 5min, a second 5min (10min),. cndot.,. 5min (25min), a sixth 5min (30min)
Secondly, obtaining a first congestion state corresponding to a first sub-target prediction time period by using a congestion state prediction model according to sensor data before the current time;
the congestion state after a preset time period, namely 5min, can be obtained based on the congestion state model according to the sensor data before the current time, namely the first congestion state corresponding to the first sub-target prediction time period.
Thirdly, obtaining a second congestion state corresponding to a second sub-prediction time period by using the congestion state prediction model again according to the sensor data before the current time, the first sub-prediction target time period and the corresponding first congestion state;
and integrating the first sub-prediction target time interval and the corresponding first congestion state (congestion state after 5min) into the sensor data before the current time, and obtaining the congestion state after 10min, namely the second congestion state corresponding to the second sub-prediction target time interval, by using the congestion state prediction model again.
And fourthly, circularly utilizing the congestion state prediction model according to the mode of obtaining the second congestion state to sequentially obtain the congestion states corresponding to the residual sub-target prediction time periods.
And integrating the second sub-prediction target time interval, the corresponding second congestion state (congestion state after 10min) and the previous integration result, and obtaining the congestion state after 15min, namely a third congestion state corresponding to the third sub-prediction target time interval, by using the congestion state prediction model again. In this way, the fourth congestion state, the fifth congestion state and the sixth congestion state can be obtained in sequence. Assuming that the congestion status is represented by a number from 0 to 3: unobstructed-0, slightly congested-1, moderately congested-2, and heavily congested-3, one form of the congestion state prediction result in the target prediction time period may be obtained:
congestion status after 30 min-sensor ID: 0,1,2,3,3,2
The above numbers represent the congestion state after 5min, 10min, 15min, 20min, 25min and 30min, respectively.
As can be seen from the process of predicting the congestion state in the target prediction time period, the principle of prediction is to cyclically utilize the congestion state prediction model to perform prediction to obtain the prediction result of the congestion state in the target prediction time period, and the number of cycles is equal to the multiple of the integral multiple.
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 predicting congestion status, which implements the method described in fig. 1 to 2, as shown in fig. 3, the apparatus includes:
the first acquiring unit 31 is configured to acquire sensor data before a current time, where the sensor data is data acquired by a sensor installed on each road of a toll station;
the prediction unit 32 is configured to perform congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a predetermined time, where the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree.
As can be seen from the above description, in the congestion status prediction apparatus according to the embodiment of the present application, first, sensor data before the current time is obtained, where the sensor data is data acquired by a sensor installed on each road of a toll station; and performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a preset time, wherein the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree. The method and the device are applied to a road monitoring system of a toll station, can predict the road congestion state in advance according to the sensor data acquired by the installed sensors of all roads, and can adjust the roads and mobilize personnel in time, so that the utilization rate of a toll channel is improved. In addition, when prediction is carried out, prediction is carried out according to a machine algorithm model, and the accuracy and the speed of a prediction result are higher.
Further, as shown in fig. 4, the apparatus further includes:
a second obtaining unit 33, configured to obtain a target prediction time interval, where the target prediction time interval is a time interval selected by a user and required to predict a road congestion state of a toll station, and the target prediction time interval is an integral multiple of a predetermined time length;
the prediction unit 32 is further configured to perform prediction by cyclically using the congestion state prediction model to obtain a congestion state prediction result in a target prediction time period, where the number of times of the cycle is equal to a multiple of the integer multiple.
Further, as shown in fig. 4, the apparatus further includes:
a third acquiring unit 34 configured to acquire sensor sample data;
the sample feature generating unit 35 is configured to preprocess the sensor sample data and generate feature data corresponding to the sensor sample data;
and the model generating unit 36 is configured to train feature data corresponding to the sensor sample data based on an integrated machine learning algorithm of the decision tree to obtain a congestion state prediction model.
Further, as shown in fig. 4, the sample feature generating unit 35 includes:
the first feature generation module 351 is configured to preprocess sensor sample data, where each piece of data in the sensor sample data obtains corresponding first feature data, and the first feature data includes one or more features of a sensor identity ID, time feature data, current congestion state data, and road feature data;
the second feature generation module 352 is configured to generate second feature data corresponding to each piece of data according to the congestion state before the time corresponding to each piece of data, where the second feature data includes one or more features of the congestion state at consecutive time points, the accumulated congestion state at consecutive time points, and the accumulated congestion state at interval time points.
Further, as shown in fig. 4, the prediction unit 32 further includes:
a time interval dividing module 321, configured to equally divide the target prediction time interval into multiple sub-target prediction time intervals according to a predetermined time length;
the prediction module 322 is configured to obtain a first congestion state corresponding to a first sub-target prediction time period by using a congestion state prediction model according to sensor data before a current time;
the prediction module 322 is further configured to obtain a second congestion state corresponding to a second sub-prediction target time period by using the congestion state prediction model again according to the sensor data before the current time, the first sub-prediction target time period, and the corresponding first congestion state;
the prediction module 322 is further configured to sequentially acquire the congestion states corresponding to all remaining sub-target prediction time periods by cyclically utilizing the congestion state prediction model according to the manner of acquiring the second congestion state.
Further, as shown in fig. 4, the apparatus further includes:
and the feature determination unit 37 is configured to, before a congestion state prediction is performed based on the congestion state prediction model according to the sensor data and a prediction result of a corresponding congestion state after a predetermined time period is obtained, pre-process the sensor data before the current time and generate feature data corresponding to the sensor data before the current time.
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.
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 predicting congestion status 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 congestion status prediction 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 (10)
1. A method of congestion status prediction, the method comprising:
acquiring sensor data before the current moment, wherein the sensor data is data acquired by a sensor arranged on each road of a toll station;
and performing congestion state prediction based on a congestion state prediction model according to the sensor data to obtain a prediction result of a congestion state corresponding to a road where the sensor is located after a preset time, wherein the congestion state prediction model is obtained by training sensor sample data based on an integrated machine learning algorithm of a decision tree.
2. The method of congestion status prediction according to claim 1, characterized in that the method further comprises:
acquiring a target prediction time interval, wherein the target prediction time interval is a time interval which is selected by a user and needs to predict the road congestion state of a toll station, and the target prediction time interval is an integral multiple of a preset time length;
and circularly utilizing the congestion state prediction model to carry out prediction to obtain a congestion state prediction result in a target prediction time period, wherein the number of circulation is equal to the multiple of the integral multiple.
3. The method of congestion status prediction according to claim 1 or 2, characterized in that the method further comprises:
acquiring sensor sample data;
preprocessing the sensor sample data and generating characteristic data corresponding to the sensor sample data;
and training feature data corresponding to the sensor sample data by an integrated machine learning algorithm based on a decision tree to obtain a congestion state prediction model.
4. The method of predicting congestion status according to claim 3, wherein the preprocessing sensor sample data and generating feature data corresponding to the sensor sample data comprises:
preprocessing sensor sample data, wherein each piece of data in the sensor sample data obtains corresponding first characteristic data, and the first characteristic data comprises one or more characteristics of a sensor Identity (ID), time characteristic data, current congestion state data and road characteristic data; and the number of the first and second electrodes,
and generating second characteristic data corresponding to each piece of data according to the congestion state before the time corresponding to each piece of data, wherein the second characteristic data comprises one or more characteristics of the congestion state at continuous time points, the accumulated congestion state at continuous time points and the accumulated congestion state at interval time points.
5. The method of predicting congestion status according to claim 2, wherein the cyclically using the congestion status prediction model to perform prediction to obtain the prediction result of congestion status in the target prediction period comprises:
equally dividing the target prediction time interval into a plurality of sub-target prediction time intervals according to a preset time length;
obtaining a first congestion state corresponding to a first sub-target prediction time period by using a congestion state prediction model according to sensor data before the current time;
obtaining a second congestion state corresponding to a second sub-prediction target time period by using the congestion state prediction model again according to the sensor data before the current time, the first sub-prediction target time period and the corresponding first congestion state;
and sequentially acquiring the congestion states corresponding to the residual sub-target prediction time periods by circularly utilizing the congestion state prediction model according to the mode of acquiring the second congestion state.
6. The method of congestion status prediction according to claim 4, wherein before performing congestion status prediction based on a congestion status prediction model based on said sensor data to obtain a prediction result of a corresponding congestion status after a predetermined period of time, said method further comprises:
and preprocessing the sensor data before the current moment and generating characteristic data corresponding to the sensor data before the current moment.
7. An apparatus for congestion status prediction, the apparatus comprising:
the system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring sensor data before the current moment, and the sensor data is data acquired by a sensor arranged on each road of a toll station;
and the prediction unit is used for predicting the congestion state based on a congestion state prediction model according to the sensor data to obtain a prediction result of the congestion state corresponding to the road where the sensor is located after a preset time, and the congestion state prediction model is obtained by training the sensor sample data based on an integrated machine learning algorithm of a decision tree.
8. The congestion status prediction apparatus according to claim 7, further comprising:
the second acquisition unit is used for acquiring a target prediction time interval, wherein the target prediction time interval is a time interval which is selected by a user and needs to predict the road congestion state of the toll station, and the target prediction time interval is an integral multiple of a preset time length;
the prediction unit is further configured to perform prediction by cyclically using the congestion state prediction model to obtain a congestion state prediction result in a target prediction time period, and the number of times of the cycle is equal to a multiple of the integral multiple.
9. A computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method for congestion status prediction according to any one of claims 1 to 6.
10. 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 congestion status prediction according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010100665.5A CN111341098B (en) | 2020-02-18 | 2020-02-18 | Congestion state prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010100665.5A CN111341098B (en) | 2020-02-18 | 2020-02-18 | Congestion state prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111341098A true CN111341098A (en) | 2020-06-26 |
| CN111341098B CN111341098B (en) | 2021-07-27 |
Family
ID=71183856
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010100665.5A Active CN111341098B (en) | 2020-02-18 | 2020-02-18 | Congestion state prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111341098B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111757272A (en) * | 2020-06-29 | 2020-10-09 | 北京百度网讯科技有限公司 | Subway congestion degree prediction method, model training method and device |
| CN112422443A (en) * | 2020-10-27 | 2021-02-26 | 鹏城实验室 | Adaptive control method, storage medium, equipment and system of congestion algorithm |
| CN112785841A (en) * | 2020-12-25 | 2021-05-11 | 北京中交兴路信息科技有限公司 | Method and device for judging congestion state of truck driving route |
| CN116432803A (en) * | 2022-12-13 | 2023-07-14 | 陕西高速电子工程有限公司 | Communication data intelligent early warning system and method based on Internet of things |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH01144200A (en) * | 1987-11-30 | 1989-06-06 | Omron Tateisi Electron Co | Traffic flow investigation system for automobile on road for exclusive automobile use |
| CA2290301A1 (en) * | 1999-03-05 | 2000-09-05 | Loran Network Management Ltd. | A method for detecting outlier measures of activity |
| CN1576789A (en) * | 2003-07-30 | 2005-02-09 | 日本先锋公司 | Device, system, method and program for notifying traffic condition and recording medium storing the program |
| CN101438334A (en) * | 2006-03-03 | 2009-05-20 | 因瑞克斯有限公司 | Dynamic time series prediction of future traffic conditions |
| CN101681553A (en) * | 2007-08-22 | 2010-03-24 | 丰田自动车株式会社 | Traffic information creation method, traffic information creation device, and navigation system |
| CN103839409A (en) * | 2014-02-27 | 2014-06-04 | 南京大学 | Traffic flow state judgment method based on multiple-cross-section vision sensing clustering analysis |
| CN105788249A (en) * | 2014-12-16 | 2016-07-20 | 高德软件有限公司 | Traffic flow prediction method, prediction model generation method and device |
| CN106781464A (en) * | 2016-12-06 | 2017-05-31 | 广州市科恩电脑有限公司 | A kind of congestion in road situation method of testing |
| CN108022425A (en) * | 2017-12-21 | 2018-05-11 | 东软集团股份有限公司 | Traffic movement prediction method, device and computer equipment |
| CN108091131A (en) * | 2016-11-22 | 2018-05-29 | 中国移动通信有限公司研究院 | A kind of traffic events recognition methods and device |
| CN108280998A (en) * | 2018-01-30 | 2018-07-13 | 浙江大学 | Short-time Traffic Flow Forecasting Methods based on historical data dynamic select |
| CN108364471A (en) * | 2018-04-24 | 2018-08-03 | 浙江方大智控科技有限公司 | Freight planning management method and system based on intelligent traffic light OD information inspections |
| CN109255948A (en) * | 2018-08-10 | 2019-01-22 | 昆明理工大学 | A kind of divided lane wagon flow scale prediction method based on Kalman filtering |
| CN109887283A (en) * | 2019-03-07 | 2019-06-14 | 东莞数汇大数据有限公司 | A kind of congestion in road prediction technique, system and device based on bayonet data |
-
2020
- 2020-02-18 CN CN202010100665.5A patent/CN111341098B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH01144200A (en) * | 1987-11-30 | 1989-06-06 | Omron Tateisi Electron Co | Traffic flow investigation system for automobile on road for exclusive automobile use |
| CA2290301A1 (en) * | 1999-03-05 | 2000-09-05 | Loran Network Management Ltd. | A method for detecting outlier measures of activity |
| CN1576789A (en) * | 2003-07-30 | 2005-02-09 | 日本先锋公司 | Device, system, method and program for notifying traffic condition and recording medium storing the program |
| CN101438334A (en) * | 2006-03-03 | 2009-05-20 | 因瑞克斯有限公司 | Dynamic time series prediction of future traffic conditions |
| CN101681553A (en) * | 2007-08-22 | 2010-03-24 | 丰田自动车株式会社 | Traffic information creation method, traffic information creation device, and navigation system |
| CN103839409A (en) * | 2014-02-27 | 2014-06-04 | 南京大学 | Traffic flow state judgment method based on multiple-cross-section vision sensing clustering analysis |
| CN105788249A (en) * | 2014-12-16 | 2016-07-20 | 高德软件有限公司 | Traffic flow prediction method, prediction model generation method and device |
| CN108091131A (en) * | 2016-11-22 | 2018-05-29 | 中国移动通信有限公司研究院 | A kind of traffic events recognition methods and device |
| CN106781464A (en) * | 2016-12-06 | 2017-05-31 | 广州市科恩电脑有限公司 | A kind of congestion in road situation method of testing |
| CN108022425A (en) * | 2017-12-21 | 2018-05-11 | 东软集团股份有限公司 | Traffic movement prediction method, device and computer equipment |
| CN108280998A (en) * | 2018-01-30 | 2018-07-13 | 浙江大学 | Short-time Traffic Flow Forecasting Methods based on historical data dynamic select |
| CN108364471A (en) * | 2018-04-24 | 2018-08-03 | 浙江方大智控科技有限公司 | Freight planning management method and system based on intelligent traffic light OD information inspections |
| CN109255948A (en) * | 2018-08-10 | 2019-01-22 | 昆明理工大学 | A kind of divided lane wagon flow scale prediction method based on Kalman filtering |
| CN109887283A (en) * | 2019-03-07 | 2019-06-14 | 东莞数汇大数据有限公司 | A kind of congestion in road prediction technique, system and device based on bayonet data |
Non-Patent Citations (3)
| Title |
|---|
| 房丽侠等: "基于决策树的交通拥挤状态预测", 《河北工业大学学报》 * |
| 杜长海: "基于神经网络和Markov链的交通流实时滚动预测", 《系统仿真学报》 * |
| 陈俊刚: "基于梯度提升决策树的交通流量预测方法研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111757272A (en) * | 2020-06-29 | 2020-10-09 | 北京百度网讯科技有限公司 | Subway congestion degree prediction method, model training method and device |
| CN112422443A (en) * | 2020-10-27 | 2021-02-26 | 鹏城实验室 | Adaptive control method, storage medium, equipment and system of congestion algorithm |
| CN112785841A (en) * | 2020-12-25 | 2021-05-11 | 北京中交兴路信息科技有限公司 | Method and device for judging congestion state of truck driving route |
| CN112785841B (en) * | 2020-12-25 | 2022-05-06 | 北京中交兴路信息科技有限公司 | Method and device for judging congestion state of truck driving route |
| CN116432803A (en) * | 2022-12-13 | 2023-07-14 | 陕西高速电子工程有限公司 | Communication data intelligent early warning system and method based on Internet of things |
| CN116432803B (en) * | 2022-12-13 | 2023-09-15 | 陕西高速电子工程有限公司 | Communication data intelligent early warning system and method based on Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111341098B (en) | 2021-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111341098B (en) | Congestion state prediction method and device | |
| CN109872535B (en) | Intelligent traffic passage prediction method, device and server | |
| CN108197739B (en) | Urban rail transit passenger flow prediction method | |
| CN107038478B (en) | Road condition prediction method and device, computer equipment and readable medium | |
| CN110264735A (en) | A kind of traffic congestion forecasting system, method and storage medium based on big data | |
| CN107862863B (en) | Method and device for improving traffic data quality | |
| EP4080408A1 (en) | Model generation method and apparatus, object detection method and apparatus, device, and storage medium | |
| CN111652280A (en) | Behavior-based target object data analysis method and device and storage medium | |
| CN110110339B (en) | Japanese hydrologic forecast error correction method and system | |
| CN111815973B (en) | Signalized intersection analysis method and related equipment | |
| CN114066236A (en) | Wind and solar active power output scene generation method and device, electronic equipment and storage medium | |
| CN110543699B (en) | Shared vehicle travel data simulation and shared vehicle scheduling method, device and equipment | |
| CN115081920A (en) | Attendance check attendance check scheduling management method, system, equipment and storage medium | |
| CN114116742A (en) | Method and device for filling time sequence data based on subway comprehensive monitoring system | |
| CN111325572B (en) | Data processing method and device | |
| CN112652164A (en) | Traffic time interval dividing method, device and equipment | |
| CN117130888A (en) | Automatic server tuning method and system based on historical data and machine learning | |
| CN111160594B (en) | Method and device for estimating arrival time and storage medium | |
| CN116956155A (en) | Sewage treatment intelligent dosing method and system based on artificial intelligence | |
| CN116611812A (en) | Artificial intelligence-based vehicle accessory damage assessment method, device, equipment and medium | |
| CN111950753A (en) | Scenic spot passenger flow prediction method and device | |
| CN114331299B (en) | Data processing method and device, electronic equipment and computer readable storage medium | |
| CN111489171B (en) | Riding travel matching method and device based on two-dimensional code, electronic equipment and medium | |
| CN113807703A (en) | Driving behavior scoring method and device, electronic equipment and medium | |
| CN111339156B (en) | Method, apparatus and computer readable storage medium for long-term determination of business data |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220513 Address after: 201306 C Building, 888 Huanhu West Second Road, Pudong New Area, Shanghai Patentee after: Zhonghaifu Information Technology Co.,Ltd. Address before: Room 302-1, floor 3, building 3, yard 128, South Fourth Ring West Road, Fengtai District, Beijing 100070 Patentee before: Beijing Zhongyi new energy Co.,Ltd. |
|
| TR01 | Transfer of patent right |