CN111325983B - Intelligent networking automobile power demand online prediction method based on space-time cooperation - Google Patents
Intelligent networking automobile power demand online prediction method based on space-time cooperation Download PDFInfo
- Publication number
- CN111325983B CN111325983B CN202010170149.XA CN202010170149A CN111325983B CN 111325983 B CN111325983 B CN 111325983B CN 202010170149 A CN202010170149 A CN 202010170149A CN 111325983 B CN111325983 B CN 111325983B
- Authority
- CN
- China
- Prior art keywords
- space
- target
- matrix
- node
- under
- 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
- 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
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
-
- 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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an intelligent networking automobile power demand online prediction method based on space-time coordination, which is characterized in that under an intelligent networking environment, a vehicle target travel is divided into a plurality of spatial road sections according to spatial distance, vehicle characteristic information of each spatial road section at a typical moment is extracted through V2I communication, a characteristic parameter spatial matrix is formed, a characteristic parameter recombination model is constructed, and an optimal node matrix set is further determined; in actual driving, the vehicle extracts vehicle characteristic information of each space road section at the current moment, calculates the power requirement of the target vehicle on the space road section through an optimal node matrix set determined by the model, and then compensates the power requirement of the space road section through a time deviation coefficient matrix to determine the final power requirement. The method can improve the power demand prediction precision of the target vehicle under the target travel and ensure the space-time cooperativity, thereby laying a foundation for the implementation of the energy management strategy and the planning of the optimal path of the intelligent networked automobile.
Description
Technical Field
The invention belongs to the field of intelligent networked automobiles, and particularly relates to an intelligent networked automobile power demand online prediction method based on space-time cooperation;
background
With the development of the 5G technology, intelligent networked automobiles gradually appear in the visual field of people and are concerned by more and more researchers, wherein the intelligent networked automobiles refer to the organic combination of the internet of vehicles and intelligent automobiles, are provided with advanced vehicle-mounted sensors, controllers, actuators and other devices, integrate modern communication and network technologies, realize the exchange and sharing of intelligent information such as automobiles, people, automobiles, roads, backgrounds and the like, realize safe, comfortable, energy-saving and efficient driving, and finally can replace people to operate;
under the intelligent networking environment, a target vehicle can grasp traffic information of other vehicles in a route in real time, exchange of vehicle information is achieved, and the method is favored by vast research students, wherein the power demand of the target vehicle on the target route is a research hotspot of an intelligent networking automobile, but the prediction method of the power demand of the intelligent networking automobile at the present stage has certain defects, such as road sections are not divided into the target route, and the change condition of the traffic volume of the target road sections cannot be accurately grasped; a proper processing method is not established for the traffic volume information of the road section; the key parameters for establishing the model are not clear, so that the prediction accuracy cannot be achieved; the space-time deviation is not considered, so that the power prediction of the target road section at different moments has certain deviation; when the method is applied to power demand prediction of the intelligent networked automobile, the prediction precision is inevitably very low, and accurate prediction cannot be achieved, so that how to accurately predict the power demand on line by using the traffic information of the target journey becomes a problem which needs to be solved urgently, the accuracy of power demand prediction is important to the implementation of the following energy management strategy and the optimal path planning of the intelligent networked automobile, and the method has important theoretical significance and research value.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an intelligent networking automobile power demand online prediction method based on space-time collaboration so as to improve the power demand prediction precision of a target vehicle in a target travel on the premise of ensuring the space-time collaboration, thereby providing an important basis for the implementation of an energy management strategy and the planning of an optimal path of an intelligent networking automobile.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention relates to an intelligent networking automobile power demand online prediction method based on space-time coordination, which is characterized by comprising the following steps of:
under the environment of intelligent networking, each group of target trips in M groups of target trips is divided into n space road sections according to space distance, and typical time t of a target vehicle under the M groups of target trips is extracted through V2I communication0The vehicle characteristic information of (1), then any ith group itemThe characteristic parameter space matrix formed by the characteristic information of the vehicle under the standard journey is Xi=(Xi1,...,Xik,...,Xin) Wherein X isikThe characteristic parameter space matrix which is composed of n-k +1 space sections from the k space section to the n space section under the ith group of target travel is represented, and the characteristic parameter space matrix comprises: xik=[xik,xi(k+1),…,xin]Wherein x isik=(sik,aik,vik,θik,dik)TIs a characteristic parameter space vector, s, of the k space section under the ith group of target travelikIs the position of the vehicle in the k space section under the ith group of target travel, aikVehicle acceleration, v, for the k-th space segment at the i-th set of target journeysikIs the vehicle speed, theta, of the k-th space segment at the i-th set of target tripsikRoad grade, d, for the k-th space segment at the i-th set of target journeysikI is more than or equal to 1 and less than or equal to M, and k is more than or equal to 1 and less than or equal to n;
defining the actual demand power vector P of n space road sections of the target vehicle under the ith group of target traveli=(Pi1,Pi2,...,Pik,...,Pin) Wherein P isikRepresenting the actual required power of the k space section of the target vehicle under the ith group of target routes;
step 2, constructing a characteristic parameter recombination model;
step 2.1, defining a characteristic parameter recombination model;
defining the characteristic parameter recombination model to be composed of an input layer, a recombination layer, a characteristic output layer and a power output layer;
defining the input layer to be composed of n-k +1 nodes, the reconstruction layer to be composed of n-k +1 nodes, wherein each node is composed of a characteristic coefficient delta and an offset value lambda, the characteristic output layer is composed of 1 node, and the power output layer is composed of 1 node;
n-k +1 nodes in the input layer defining the kth space segment under the ith group of target travelCharacteristic parameter space matrix X'ik=[x′i1,x′i2,…,x′ij,...,x′i(n-k+1)]WhereinRespectively representing characteristic parameter space vectors corresponding to the 1 st node, the 2 nd node and the (n-k + 1) th node in the input layer of the kth space section under the ith group of target travel;
defining a node matrix consisting of n-k +1 nodes in a reconstruction layer of a k space section under the ith group of target travelWherein the content of the first and second substances, respectively is the characteristic coefficient and the deviant of the jth node in the reconstruction layer of the kth space section under the ith group of target travel;
defining a node output value matrix Y corresponding to n-k +1 nodes in a reconstruction layer of a k-th space road section under the ith group of target travelik=[yi1,yi2,...,yij,...,yi(n-k+1)]Wherein, yijIs the node output value corresponding to the jth node in the reconstruction layer of the kth space road section under the ith group of target travel;
step 2.2, initializing i to 1;
step 2.3, initializing k to be 1;
step 2.4, calculating the weighted association coefficient gamma of the jth node in the reconstruction layer of the kth space section under the ith group of target travel by using the formula (1) and the formula (2)ijAnd a weighting ratio xiij;
Step 2.5, determining a recombination layer node matrix set under the ith group of target strokes;
step 2.5.1, calculating a node output value y corresponding to the jth node in the reconstruction layer of the kth space section under the ith group of target travel by using the formula (3)ijSo as to obtain a node output value matrix Y corresponding to n-k +1 nodes in the reconstruction layer of the kth space road section under the ith group of target travelik=[yi1,yi2,...,yij,...,yi(n-k+1)];
In the formula (3), the reaction mixture is,representing the sum of the characteristic parameter space vectors corresponding to n-k +1 nodes in the input layer of the kth space section under the ith group of target travel;
step 2.5.2, calculating a variable weighting recombination function J of the kth space section under the ith group of target travel by using the formula (4)ik(δ,λ);
Step 2.5.3, definition of JikThe upper limit of (delta, lambda) is epsilon, if JikWhen (delta, lambda) is less than or equal to epsilon, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travelAnd offset valueKeeping unchanged, and sequentially executing the step 2.5.4;
if JikWhen the (delta, lambda) > epsilon, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travel is obtained by using the formula (5) for correctionAnd offset valueAnd respectively assigned withAndthen step 2.5.1 is executed;
in the formula (5), μ and ν are correction rates;
step 2.5.4, after k +1 is assigned with k, whether k is more than n is judged, if yes, the recombination layer node matrix set χ of n space road sections under the ith group of target travel is obtainedi=[χi1,χi2,...,χik,...,χin]WhereinA node matrix composed of n-k +1 nodes in a reconstruction layer of a k-th space section under the ith group of target travel, and comprising: respectively is the characteristic coefficient and the deviant of the jth node in the reconstruction layer of the kth space section under the ith group of target travel; otherwise, returning to the step 2.4;
step 2.6, calculating the kth target vehicle in the ith group of target travel by using the formula (6)Characteristic parameter recombination matrix T of space road sectionikSo as to obtain a characteristic parameter recombination matrix set T of n space road sections of the target vehicle under the ith group of target traveli=(Ti1,Ti2,...,Tik,...,Tin);
Step 2.7, calculating power demand P 'of k space road section of the target vehicle under the i set of target journey by using formula (7)'ikObtaining power demand vectors P 'of n space road segments of the target vehicle under the ith group of target routes'i=(P′i1,P′i2,...,P′ik,...,P′in);
In the formula (7), m is the mass of the target vehicle,is a characteristic parameter recombination matrix TikThe road slope vector of (1) is,reorganizing the matrix T for the characteristic parametersikThe speed vector of the vehicle (2) is,reorganizing the matrix T for the characteristic parametersikThe vector of the medium resistance force is,reorganizing the matrix T for the characteristic parametersikThe rolling resistance coefficient vector of (1) is,reorganizing the matrix T for the characteristic parametersikAn acceleration vector of (1);
step 2.8, calculating the power deviation value E (P) of the n space road sections of the target vehicle under the ith group of target travel by using the formula (8)i,P′i);
Step 2.9, definition of E (P)i,P′i) Has an upper limit of σ, if E (P)i,P′i) If not more than sigma, executing step 2.10;
if E (P)i,P′i) If the distance is larger than sigma, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travel is obtained by using the formula (9) for correctionAnd offset valueAnd respectively assigned withAndthen step 2.6 is executed;
in the formula (9), μ ', ν' is a correction rate;
step 2.10, after i +1 is assigned to i, whether i is greater than M or not is judged, if yes, node matrix sets χ of n space road sections under the Mth group of target travelM=[χΜ1,χM2,...,χMk,...,χMn]As a set of optimal node matrices, wherein,indicating the kth empty in the Mth group of target tripsThe node matrix formed by n-k +1 nodes in the reconstruction layer of the inter-road section comprises the following components: respectively obtaining a characteristic coefficient and an offset value of a jth node in a reconstruction layer of a kth space section under the Mth group of target strokes; otherwise, returning to the step 2.4;
step 3, calculating the time deviation coefficient alpha of the kth space section at phi moment by using the formula (10)φkSo as to obtain the time deviation coefficient matrix of n space road sections under phi moment
In equation (10), φ is the time before the target vehicle enters the target course, t0A typical time when the target vehicle travels in the target trip, L is a link length of the target trip, n is the number of spatial links dividing the target trip,the running speed of the target vehicle on the spatial road section j is shown, and w is the running speed of the target vehicle on the spatial road section t0The rotational speed of the engine at the moment,a time of travel for the target vehicle from the 1 st spatial segment to the k-th spatial segment;
step 4, calculating the target vehicle in A by using the formula (11)φFinal power demand column vector P for next n spatial road segmentsφ=(Pφ1,Pφ2,...,Pφk,...,Pφn)T;
Pφ=Aφ×P′ (11)
In formula (11), P '═ P'1,P′2,...,P′k,...,P′n)TThe power demand column vector of the target vehicle in n space road sections predicted by the characteristic parameter recombination model, AφThe time deviation coefficient matrix of n space road sections at phi moment, and T represents the transposition of the vector;
and 5, in an actual target travel, a target vehicle acquires vehicle driving information of n spatial road sections in the target travel at the current moment in advance through V2I communication, a characteristic parameter spatial matrix of the spatial road sections is formed, then a formula (6) is used for calculating a characteristic parameter recombination matrix set of the n spatial road sections by using the optimal node matrix set determined in the step 2, a formula (7) is used for obtaining power demand vectors of the n spatial road sections predicted by the characteristic parameter recombination model, a formula (10) is used for obtaining time deviation coefficient matrices of the n spatial road sections, and finally a formula (11) is used for calculating the final power demand vector of the target vehicle on the n spatial road sections.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention designs an intelligent networking automobile power demand online prediction method based on space-time coordination, which can be used for acquiring vehicle information of each space road section of a vehicle in a target travel in advance through a V2I communication technology in an intelligent networking environment to form a characteristic parameter space matrix, and calculating the power demand of each space road section by combining with an optimal node matrix set determined by a characteristic parameter recombination model, and considering the same problem of space-time coordination, so that the power demand online prediction is more accurate.
2. The invention provides a method for dividing a target travel, which is characterized in that the target travel is divided into a plurality of spatial road sections according to spatial distance, then traffic information of each spatial road section is extracted to form a characteristic parameter spatial matrix of the target travel, the grasp of vehicle flow and road conditions of the whole line is improved, a foundation is laid for the establishment of a characteristic parameter recombination model, and the prediction accuracy of the characteristic parameter recombination model is ensured;
3. the invention provides a characteristic parameter recombination model capable of self-training, which consists of two layers of cycles, wherein the outer cycle is controlled by the group number i of a target travel, the inner cycle is controlled by the positions k of n space road sections, the recombination layer is the key point of the model, a weighting correlation coefficient and a weighting ratio are provided, characteristic parameter vectors influencing all space road sections of a target vehicle in front of the current space road section are endowed with different weights, the output value of each node of the recombination layer is a linear combination of the characteristic parameter vectors of the current front space road section with different weights, then the characteristic coefficient and the offset value of each node of the recombination layer are determined, and an optimal node matrix is further determined.
4. The power has space-time cooperativity in prediction, and the invention introduces a time deviation coefficient matrix AφThe space-time difference of the space road section block power prediction is solved, the accurate prediction of the power requirements of the target vehicle on n space road sections at the current moment is realized, meanwhile, preconditions are provided for the implementation of an energy management strategy and the planning of an optimal path of the intelligent networked automobile, and the method has a wide application prospect.
Drawings
FIG. 1 is a flow chart of the present invention for building a feature parameter reorganization model;
FIG. 2 is a basic flow chart of the power demand online prediction method of the present invention.
Detailed Description
In the embodiment, an intelligent networking automobile power demand online prediction method based on space-time coordination is characterized in that under an intelligent networking environment, a vehicle target travel is divided into a plurality of spatial road sections according to spatial distance, vehicle characteristic information of each spatial road section at a typical moment is extracted through V2I communication, a characteristic parameter spatial matrix is formed, a characteristic parameter recombination model is constructed, and an optimal node matrix set is further determined; in actual driving, the vehicle extracts vehicle characteristic information of each space road section at the current moment, calculates the power requirement of the target vehicle on the space road section through an optimal node matrix set determined by the model, and then compensates the power requirement of the space road section through a time deviation coefficient matrix to determine the final power requirement. As shown in fig. 2, specifically, the following steps are performed:
under the environment of intelligent networking, each group of target trips in M groups of target trips is divided into n space road sections according to space distance, and typical time t of a target vehicle under the M groups of target trips is extracted through V2I communication0The characteristic parameter space matrix formed by the vehicle characteristic information under any ith group of target travel is Xi=(Xi1,...,Xik,...,Xin) Wherein X isikThe characteristic parameter space matrix which is composed of n-k +1 space sections from the k space section to the n space section under the ith group of target travel is represented, and the characteristic parameter space matrix comprises: xik=[xik,xi(k+1),…,xin]Wherein x isik=(sik,aik,vik,θik,dik)TIs a characteristic parameter space vector, s, of the k space section under the ith group of target travelikIs the position of the vehicle in the k space section under the ith group of target travel, aikVehicle acceleration, v, for the k-th space segment at the i-th set of target journeysikIs the vehicle speed, theta, of the k-th space segment at the i-th set of target tripsikRoad grade, d, for the k-th space segment at the i-th set of target journeysikI is more than or equal to 1 and less than or equal to M, and k is more than or equal to 1 and less than or equal to n;
defining the actual demand power vector P of n space road sections of the target vehicle under the ith group of target traveli=(Pi1,Pi2,...,Pik,...,Pin) Wherein P isikRepresenting the actual required power of the k space section of the target vehicle under the ith group of target routes;
in this embodiment, M-20 sets of target trips are selected, and each set of target trips is divided into n-10 spatial road segments, t0If the vehicle characteristic information in the target journey at the time of 10 is extracted, the vehicle characteristic information under the ith group of target journeys is constructedThe characteristic parameter space matrix is Xi=(Xi1,Xi2,Xi3,Xi4,Xi5,Xi6,Xi7,Xi8,Xi9,Xi10),1≤i≤20;
Step 2, constructing a characteristic parameter recombination model;
step 2.1, defining a characteristic parameter recombination model;
defining the characteristic parameter recombination model to be composed of an input layer, a recombination layer, a characteristic output layer and a power output layer;
defining the input layer to be composed of n-k +1 nodes, the reconstruction layer to be composed of n-k +1 nodes, wherein each node is composed of a characteristic coefficient delta and an offset value lambda, the characteristic output layer is composed of 1 node, and the power output layer is composed of 1 node;
defining a characteristic parameter space matrix X 'consisting of n-k +1 nodes in an input layer of a k space section under an i group of target strokes'ik=[x′i1,x′i2,…,x′ij,...,x′i(n-k+1)]WhereinRespectively representing characteristic parameter space vectors corresponding to the 1 st node, the 2 nd node and the (n-k + 1) th node in the input layer of the kth space section under the ith group of target travel;
defining a node matrix consisting of n-k +1 nodes in a reconstruction layer of a k space section under the ith group of target travelWherein the content of the first and second substances, respectively is the characteristic coefficient and the deviant of the jth node in the reconstruction layer of the kth space section under the ith group of target travel;
defining a node output value matrix Y corresponding to n-k +1 nodes in a reconstruction layer of a k-th space road section under the ith group of target travelik=[yi1,yi2,...,yij,...,yi(n-k+1)]Wherein, yijIs the node output value corresponding to the jth node in the reconstruction layer of the kth space road section under the ith group of target travel;
step 2.2, initializing i to 1;
step 2.3, initializing k to be 1;
step 2.4, calculating the weighted association coefficient gamma of the jth node in the reconstruction layer of the kth space section under the ith group of target travel by using the formula (1) and the formula (2)ijAnd a weighting ratio xiij;
Step 2.5, determining a recombination layer node matrix set under the ith group of target strokes;
step 2.5.1, calculating a node output value y corresponding to the jth node in the reconstruction layer of the kth space section under the ith group of target travel by using the formula (3)ijSo as to obtain a node output value matrix Y corresponding to n-k +1 nodes in the reconstruction layer of the kth space road section under the ith group of target travelik=[yi1,yi2,...,yij,...,yi(n-k+1)];
In the formula (3), the reaction mixture is,representing the characteristic parameter space vector corresponding to n-k +1 nodes in the input layer of the kth space section under the ith group of target travelAnd;
step 2.5.2, calculating a variable weighting recombination function J of the kth space section under the ith group of target travel by using the formula (4)ik(δ,λ);
Step 2.5.3, definition of JikThe upper limit of (delta, lambda) is epsilon, if JikWhen (delta, lambda) is less than or equal to epsilon, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travelAnd offset valueKeeping unchanged, and sequentially executing the step 2.5.4;
if JikWhen the (delta, lambda) > epsilon, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travel is obtained by using the formula (5) for correctionAnd offset valueAnd respectively assigned withAndthen step 2.5.1 is executed; in this example, let ε be 0.01
In the formula (5), μ and ν are correction rates; the function is to correct the characteristic coefficient and the offset value at a certain rate, in this embodiment, μ is 0.5, and ν is 0.5;
step 2.5.4, after k +1 is assigned with k, whether k is more than n is judged, if yes, the recombination layer node matrix set χ of n space road sections under the ith group of target travel is obtainedi=[χi1,χi2,...,χik,...,χin]WhereinA node matrix composed of n-k +1 nodes in a reconstruction layer of a k-th space section under the ith group of target travel, and comprising: respectively is the characteristic coefficient and the deviant of the jth node in the reconstruction layer of the kth space section under the ith group of target travel; otherwise, returning to the step 2.4;
step 2.6, calculating a characteristic parameter recombination matrix T of the kth space section of the target vehicle under the ith group of target travel by using the formula (6)ikSo as to obtain a characteristic parameter recombination matrix set T of n space road sections of the target vehicle under the ith group of target traveli=(Ti1,Ti2,...,Tik,...,Tin);
Step 2.7, calculating power demand P 'of k space road section of the target vehicle under the i set of target journey by using formula (7)'ikObtaining power demand vectors P 'of n space road segments of the target vehicle under the ith group of target routes'i=(P′i1,P′i2,...,P′ik,...,P′in);
In the formula (7), m is the mass of the target vehicle,is a characteristic parameter recombination matrix TikThe road slope vector of (1) is,reorganizing the matrix T for the characteristic parametersikThe speed vector of the vehicle (2) is,reorganizing the matrix T for the characteristic parametersikThe vector of the medium resistance force is,reorganizing the matrix T for the characteristic parametersikThe rolling resistance coefficient vector of (1) is,reorganizing the matrix T for the characteristic parametersikAn acceleration vector of (1);
in the formula (7), m is the mass of the target vehicle,is a characteristic parameter recombination matrix TikThe road slope vector of (1) is,reorganizing the matrix T for the characteristic parametersikMedium running speed vector, fkIs the resistance, xi, of the k space section of the target vehicle under the ith group of target strokeskThe rolling resistance coefficient of the k space section of the target vehicle under the ith group of target strokes is obtained,reorganizing the matrix T for the characteristic parametersikAn acceleration vector of (1);
step 2.8, calculating the power deviation of the target vehicle on n space road sections under the ith group of target travel by using the formula (8)Difference E (P)i,P′i);
Step 2.9, definition of E (P)i,P′i) Has an upper limit of σ, if E (P)i,P′i) If not more than sigma, executing step 2.10;
if E (P)i,P′i) If the distance is larger than sigma, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travel is obtained by using the formula (9) for correctionAnd offset valueAnd respectively assigned withAndthen step 2.6 is executed; in this embodiment, σ is taken to be 0.01;
in the formula (9), μ ', ν' is a correction rate; in the present embodiment, the correction rate is the same value as that of the formula (5);
step 2.10, after i +1 is assigned to i, whether i is greater than M or not is judged, if yes, node matrix sets χ of n space road sections under the Mth group of target travelM=[χΜ1,χM2,...,χMk,...,χMn]As a set of optimal node matrices, wherein,n-k +1 reconstruction layers for representing k-th space section under M-th group of target travelA node matrix composed of nodes, and having: respectively obtaining a characteristic coefficient and an offset value of a jth node in a reconstruction layer of a kth space section under the Mth group of target strokes; otherwise, returning to the step 2.4;
step 3, calculating the time deviation coefficient alpha of the kth space section at phi moment by using the formula (10)φkSo as to obtain the time deviation coefficient matrix of n space road sections under phi momentIn this embodiment, rounding is used to round the value of φ;
in equation (10), φ is the time before the target vehicle enters the target course, t0A typical time when the target vehicle travels in the target trip, L is a link length of the target trip, n is the number of spatial links dividing the target trip,the running speed of the target vehicle on the spatial road section j is shown, and w is the running speed of the target vehicle on the spatial road section t0The rotational speed of the engine at the moment,a time of travel for the target vehicle from the 1 st spatial segment to the k-th spatial segment;
step 4, calculating the target vehicle in A by using the formula (11)φFinal power demand column vector P for next n spatial road segmentsφ=(Pφ1,Pφ2,...,Pφk,...,Pφn)T;
Pφ=Aφ×P′ (11)
In formula (11), P '═ P'1,P′2,...,P′k,...,P′n)TThe power demand column vector of the target vehicle in n space road sections predicted by the characteristic parameter recombination model, AφThe time deviation coefficient matrix of n space road sections at phi moment, and T represents the transposition of the vector;
step 5, in an actual target travel, a target vehicle obtains vehicle driving information of n space road sections in the target travel at the current moment in advance through V2I communication, a characteristic parameter space matrix of the space road sections is formed, then an equation (6) is used for calculating a characteristic parameter recombination matrix set of the n space road sections by using an optimal node matrix set determined in the step 2, a power demand vector of the n space road sections predicted by a characteristic parameter recombination model is obtained by using an equation (7), a time deviation coefficient matrix of the n space road sections is obtained by using an equation (10), and finally a final power demand vector of the target vehicle on the n space road sections is calculated by using an equation (11);
in the embodiment, when the target vehicle enters the target road segment covered by the intelligent internet traffic system, the vehicle-mounted terminal of the target vehicle and the vehicle-mounted terminals of other vehicles on the front space road section perform information interaction through V2I, so that the driving information of the vehicle at the current moment is acquired, then, a characteristic parameter space matrix is formed, then, an optimal node matrix set determined by the model is used for respectively substituting an equation (6) to calculate a characteristic parameter recombination matrix set of 10 space road sections, an equation (7) is used for calculating power demand vectors of the target vehicle predicted by the characteristic parameter recombination model on the 10 space road sections, then rounding the current moment to determine a value of phi, obtaining a time deviation coefficient matrix of 10 space road sections at the phi moment by using an equation (10), and finally calculating a final power demand vector of the target vehicle at the 10 space road sections at the current moment by using an equation (11);
in conclusion, the method realizes the online prediction of the power demand of the intelligent networked automobile, and ensures the accuracy of the prediction and the time-space cooperativity, thereby laying a foundation for the implementation of the energy management strategy and the planning of the optimal path of the intelligent networked automobile and having wide application prospect.
Claims (1)
1. An intelligent networking automobile power demand online prediction method based on space-time coordination is characterized by comprising the following steps:
step 1, establishing a characteristic parameter space matrix of a space road section;
under the environment of intelligent networking, each group of target trips in M groups of target trips is divided into n space road sections according to space distance, and typical time t of a target vehicle under the M groups of target trips is extracted through V2I communication0The characteristic parameter space matrix formed by the vehicle characteristic information under any ith group of target travel is Xi=(Xi1,...,Xik,...,Xin) Wherein X isikThe characteristic parameter space matrix which is composed of n-k +1 space sections from the k space section to the n space section under the ith group of target travel is represented, and the characteristic parameter space matrix comprises: xik=[xik,xi(k+1),…,xin]Wherein x isik=(sik,aik,vik,θik,dik)TIs a characteristic parameter space vector, s, of the k space section under the ith group of target travelikIs the position of the vehicle in the k space section under the ith group of target travel, aikVehicle acceleration, v, for the k-th space segment at the i-th set of target journeysikIs the vehicle speed, theta, of the k-th space segment at the i-th set of target tripsikRoad grade, d, for the k-th space segment at the i-th set of target journeysikI is more than or equal to 1 and less than or equal to M, and k is more than or equal to 1 and less than or equal to n;
defining the actual demand power vector P of n space road sections of the target vehicle under the ith group of target traveli=(Pi1,Pi2,...,Pik,...,Pin) Wherein P isikRepresenting the actual required power of the k space section of the target vehicle under the ith group of target routes;
step 2, constructing a characteristic parameter recombination model;
step 2.1, defining a characteristic parameter recombination model;
defining the characteristic parameter recombination model to be composed of an input layer, a recombination layer, a characteristic output layer and a power output layer;
defining the input layer to be composed of n-k +1 nodes, the reconstruction layer to be composed of n-k +1 nodes, wherein each node is composed of a characteristic coefficient delta and an offset value lambda, the characteristic output layer is composed of 1 node, and the power output layer is composed of 1 node;
defining a characteristic parameter space matrix X 'consisting of n-k +1 nodes in an input layer of a k space section under an i group of target strokes'ik=[x′i1,x′i2,…,x′ij,...,x′i(n-k+1)]WhereinRespectively representing characteristic parameter space vectors corresponding to the 1 st node, the 2 nd node and the (n-k + 1) th node in the input layer of the kth space section under the ith group of target travel;
defining a node matrix consisting of n-k +1 nodes in a reconstruction layer of a k space section under the ith group of target travelWherein the content of the first and second substances, respectively is the characteristic coefficient and the deviant of the jth node in the reconstruction layer of the kth space section under the ith group of target travel;
defining a node output value matrix Y corresponding to n-k +1 nodes in a reconstruction layer of a k-th space road section under the ith group of target travelik=[yi1,yi2,...,yij,...,yi(n-k+1)]Wherein, yijIs the ith group under the target strokeOutputting a node corresponding to the jth node in the reconstruction layer of the k space road sections;
step 2.2, initializing i to 1;
step 2.3, initializing k to be 1;
step 2.4, calculating the weighted association coefficient gamma of the jth node in the reconstruction layer of the kth space section under the ith group of target travel by using the formula (1) and the formula (2)ijAnd a weighting ratio xiij;
Step 2.5, determining a recombination layer node matrix set under the ith group of target strokes;
step 2.5.1, calculating a node output value y corresponding to the jth node in the reconstruction layer of the kth space section under the ith group of target travel by using the formula (3)ijSo as to obtain a node output value matrix Y corresponding to n-k +1 nodes in the reconstruction layer of the kth space road section under the ith group of target travelik=[yi1,yi2,...,yij,...,yi(n-k+1)];
In the formula (3), the reaction mixture is,representing the sum of the characteristic parameter space vectors corresponding to n-k +1 nodes in the input layer of the kth space section under the ith group of target travel;
step 2.5.2, calculating a variable weighting recombination function J of the kth space section under the ith group of target travel by using the formula (4)ik(δ,λ);
Step 2.5.3, definition of JikThe upper limit of (delta, lambda) is epsilon, if JikWhen (delta, lambda) is less than or equal to epsilon, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travelAnd offset valueKeeping unchanged, and sequentially executing the step 2.5.4;
if JikWhen the (delta, lambda) > epsilon, the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target travel is obtained by using the formula (5) for correctionAnd offset valueAnd respectively assigned withAndthen step 2.5.1 is executed;
in the formula (5), μ and ν are correction rates;
step 2.5.4, after k +1 is assigned with k, whether k is more than n is judged, if yes, the recombination layer node matrix set χ of n space road sections under the ith group of target travel is obtainedi=[χi1,χi2,...,χik,...,χin]WhereinA node matrix composed of n-k +1 nodes in a reconstruction layer of a k-th space section under the ith group of target travel, and comprising: respectively is the characteristic coefficient and the deviant of the jth node in the reconstruction layer of the kth space section under the ith group of target travel; otherwise, returning to the step 2.4;
step 2.6, calculating a characteristic parameter recombination matrix T of the kth space section of the target vehicle under the ith group of target travel by using the formula (6)ikSo as to obtain a characteristic parameter recombination matrix set T of n space road sections of the target vehicle under the ith group of target traveli=(Ti1,Ti2,...,Tik,...,Tin);
Step 2.7, calculating power demand P 'of k space road section of the target vehicle under the i set of target journey by using formula (7)'ikSo as to obtain the power demand vector P of n space road sections of the target vehicle under the ith group of target traveli′=(P′i1,P′i2,...,P′ik,...,P′in);
In the formula (7), m is the mass of the target vehicle,is a characteristic parameter recombination matrix TikThe road slope vector of (1) is,reorganizing the matrix T for the characteristic parametersikThe speed vector of the vehicle (2) is,reorganizing the matrix T for the characteristic parametersikThe vector of the medium resistance force is,reorganizing the matrix T for the characteristic parametersikThe rolling resistance coefficient vector of (1) is,reorganizing the matrix T for the characteristic parametersikAn acceleration vector of (1);
step 2.8, calculating the power deviation value E (P) of the n space road sections of the target vehicle under the ith group of target travel by using the formula (8)i,Pi′);
Step 2.9, definition of E (P)i,Pi') has an upper limit of σ, if E (P)i,Pi') less than or equal to sigma, executing step 2.10;
if E (P)i,PiIf') is greater than sigma, then the characteristic coefficient of the jth node in the reconstruction layer of the kth space section under the ith group of target strokes is obtained by using the correction of the formula (9)And offset valueAnd respectively assigned withAndthen step 2.6 is executed;
in the formula (9), μ ', ν' is a correction rate;
step 2.10, after i +1 is assigned to i, whether i is greater than M or not is judged, if yes, node matrix sets χ of n space road sections under the Mth group of target travelM=[χΜ1,χM2,...,χMk,...,χMn]As a set of optimal node matrices, wherein,a node matrix composed of n-k +1 nodes in a reconstruction layer of a kth space section under the Mth group of target travel, and comprising: respectively obtaining a characteristic coefficient and an offset value of a jth node in a reconstruction layer of a kth space section under the Mth group of target strokes; otherwise, returning to the step 2.4;
step 3, calculating the time deviation coefficient alpha of the kth space section at phi moment by using the formula (10)φkSo as to obtain the time deviation coefficient matrix of n space road sections under phi moment
In equation (10), φ is the time before the target vehicle enters the target course, t0A typical time when the target vehicle travels in the target trip, L is a link length of the target trip, n is the number of spatial links dividing the target trip,the running speed of the target vehicle on the spatial road section j is shown, and w is the running speed of the target vehicle on the spatial road section t0The rotational speed of the engine at the moment,a time of travel for the target vehicle from the 1 st spatial segment to the k-th spatial segment;
step 4, calculating the target vehicle in A by using the formula (11)φFinal power demand column vector P for next n spatial road segmentsφ=(Pφ1,Pφ2,...,Pφk,...,Pφn)T;
Pφ=Aφ×P′ (11)
In formula (11), P ═ P1′,P2′,...,P′k,...,P′n)TThe power demand column vector of the target vehicle in n space road sections predicted by the characteristic parameter recombination model, AφThe time deviation coefficient matrix of n space road sections at phi moment, and T represents the transposition of the vector;
and 5, in an actual target travel, a target vehicle acquires vehicle driving information of n spatial road sections in the target travel at the current moment in advance through V2I communication, a characteristic parameter spatial matrix of the spatial road sections is formed, then a formula (6) is used for calculating a characteristic parameter recombination matrix set of the n spatial road sections by using the optimal node matrix set determined in the step 2, a formula (7) is used for obtaining power demand vectors of the n spatial road sections predicted by the characteristic parameter recombination model, a formula (10) is used for obtaining time deviation coefficient matrices of the n spatial road sections, and finally a formula (11) is used for calculating the final power demand vector of the target vehicle on the n spatial road sections.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010170149.XA CN111325983B (en) | 2020-03-12 | 2020-03-12 | Intelligent networking automobile power demand online prediction method based on space-time cooperation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010170149.XA CN111325983B (en) | 2020-03-12 | 2020-03-12 | Intelligent networking automobile power demand online prediction method based on space-time cooperation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111325983A CN111325983A (en) | 2020-06-23 |
CN111325983B true CN111325983B (en) | 2021-01-15 |
Family
ID=71171666
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010170149.XA Active CN111325983B (en) | 2020-03-12 | 2020-03-12 | Intelligent networking automobile power demand online prediction method based on space-time cooperation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111325983B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113114697B (en) * | 2021-04-21 | 2022-03-11 | 合肥工业大学 | Whole vehicle cloud test data online packaging method based on feature self-dimension reduction mark |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103606271A (en) * | 2013-11-27 | 2014-02-26 | 大连理工大学 | Method for controlling hybrid power urban buses |
CN104703142A (en) * | 2015-03-10 | 2015-06-10 | 大连理工大学 | Game theory-based vehicular networking location tracking vehicle power control method |
JP2018055208A (en) * | 2016-09-27 | 2018-04-05 | 本田技研工業株式会社 | Traffic obstruction risk displaying device |
CN110166980A (en) * | 2019-05-15 | 2019-08-23 | 南京邮电大学 | The power optimization method of distributing antenna system caching constraint under high-speed rail scene |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB201200700D0 (en) * | 2012-01-17 | 2012-02-29 | Rolls Royce Plc | Vehicle energy and power management method and system |
CN104008647B (en) * | 2014-06-12 | 2016-02-10 | 北京航空航天大学 | A kind of road traffic energy consumption quantization method based on motor-driven vehicle going pattern |
CN105528498B (en) * | 2016-01-13 | 2018-11-27 | 河南理工大学 | Net connection intelligent electric vehicle integrated modelling and integrated control method |
CN105759753B (en) * | 2016-01-25 | 2018-06-19 | 合肥工业大学 | The energy management optimal control method of hybrid vehicle based on V2X |
CN105946857B (en) * | 2016-05-16 | 2017-02-15 | 吉林大学 | Parallel plug-in hybrid electric vehicle (PHEV) energy management method based on intelligent transportation system |
US10690103B2 (en) * | 2017-09-26 | 2020-06-23 | Paccar Inc | Systems and methods for using an electric motor in predictive and automatic engine stop-start systems |
CN107862864B (en) * | 2017-10-18 | 2021-06-15 | 南京航空航天大学 | Driving condition intelligent prediction estimation method based on driving habits and traffic road conditions |
CN110505602A (en) * | 2018-05-17 | 2019-11-26 | 大唐移动通信设备有限公司 | A kind of data processing method of car networking, server and car networking system |
WO2020014131A1 (en) * | 2018-07-10 | 2020-01-16 | Cavh Llc | Connected automated vehicle highway systems and methods related to transit vehicles and systems |
CN110696815B (en) * | 2019-11-21 | 2020-10-09 | 北京理工大学 | Prediction energy management method of network-connected hybrid electric vehicle |
-
2020
- 2020-03-12 CN CN202010170149.XA patent/CN111325983B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103606271A (en) * | 2013-11-27 | 2014-02-26 | 大连理工大学 | Method for controlling hybrid power urban buses |
CN104703142A (en) * | 2015-03-10 | 2015-06-10 | 大连理工大学 | Game theory-based vehicular networking location tracking vehicle power control method |
JP2018055208A (en) * | 2016-09-27 | 2018-04-05 | 本田技研工業株式会社 | Traffic obstruction risk displaying device |
CN110166980A (en) * | 2019-05-15 | 2019-08-23 | 南京邮电大学 | The power optimization method of distributing antenna system caching constraint under high-speed rail scene |
Also Published As
Publication number | Publication date |
---|---|
CN111325983A (en) | 2020-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021103625A1 (en) | Short-term vehicle speed condition real-time prediction method based on interaction between vehicle ahead and current vehicle | |
CN111931905B (en) | Graph convolution neural network model and vehicle track prediction method using same | |
CN109017809B (en) | Energy distribution method based on cross-country working condition prediction | |
CN103914985B (en) | A kind of hybrid power passenger car following speed of a motor vehicle trajectory predictions method | |
CN109131350B (en) | Energy management method and system for hybrid electric vehicle | |
CN113538910B (en) | Self-adaptive full-chain urban area network signal control optimization method | |
CN111439260A (en) | Network-connected commercial diesel vehicle cruise running optimization control system oriented to individual requirements | |
CN106548254A (en) | A kind of Forecasting Methodology of effective parking position | |
CN106529818A (en) | Water quality evaluation prediction method based on fuzzy wavelet neural network | |
CN111554118B (en) | Dynamic prediction method and system for bus arrival time | |
CN111325983B (en) | Intelligent networking automobile power demand online prediction method based on space-time cooperation | |
CN113053120B (en) | Traffic signal lamp scheduling method and system based on iterative learning model predictive control | |
CN112477846A (en) | Intelligent networking electric automobile queue control method giving consideration to stability and energy conservation | |
CN115534929A (en) | Plug-in hybrid electric vehicle energy management method based on multi-information fusion | |
CN112950926A (en) | Urban trunk road speed prediction method based on big data and deep learning | |
CN110543967B (en) | Electric vehicle waiting time distribution short-time prediction method in network connection charging station environment | |
Hwang et al. | Comparative study on the prediction of city bus speed between LSTM and GRU | |
CN112098869A (en) | Self-adaptive electric vehicle SOC estimation method based on big data | |
CN107301266A (en) | A kind of ferric phosphate lithium cell LOC evaluation methods and system | |
CN115331460B (en) | Large-scale traffic signal control method and device based on deep reinforcement learning | |
US20230264578A1 (en) | Method for predicting energy consumption-recovery ratio of new energy vehicle, and energy saving control method and system for new energy vehicle | |
CN115115082B (en) | Long-distance expressway traffic flow prediction method based on federal learning | |
CN113610298B (en) | User travel energy consumption prediction and path recommendation method considering user travel behaviors | |
CN115359672A (en) | Traffic area boundary control method combining data driving and reinforcement learning | |
Li et al. | Road pavement identification based on acceleration signals of off-road vehicles using the batch normalized recurrent neural network |
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 |