CN116757559A - Emission reduction efficiency evaluation system suitable for green trip of public institution - Google Patents
Emission reduction efficiency evaluation system suitable for green trip of public institution Download PDFInfo
- Publication number
- CN116757559A CN116757559A CN202311050021.XA CN202311050021A CN116757559A CN 116757559 A CN116757559 A CN 116757559A CN 202311050021 A CN202311050021 A CN 202311050021A CN 116757559 A CN116757559 A CN 116757559A
- Authority
- CN
- China
- Prior art keywords
- travel
- vehicle
- data
- public
- training
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 52
- 230000009467 reduction Effects 0.000 title claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 73
- 230000009471 action Effects 0.000 claims abstract description 49
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims abstract description 48
- 229910052799 carbon Inorganic materials 0.000 claims abstract description 48
- 230000002787 reinforcement Effects 0.000 claims abstract description 36
- 238000013480 data collection Methods 0.000 claims abstract description 9
- 238000012544 monitoring process Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 18
- 239000000446 fuel Substances 0.000 claims description 12
- 238000006467 substitution reaction Methods 0.000 claims description 11
- 238000000034 method Methods 0.000 description 6
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 4
- 229910002092 carbon dioxide Inorganic materials 0.000 description 2
- 239000001569 carbon dioxide Substances 0.000 description 2
- 238000004134 energy conservation Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000003915 air pollution Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000003090 exacerbative effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/092—Reinforcement learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses an emission reduction efficiency evaluation system suitable for green travel of public institutions, which relates to the technical field of carbon emission reduction of public institutions, wherein an emission basic data of the public institutions is collected through a growth emission information collection module, a travel data collection module is set to collect a plurality of historical travel data of the public institutions in a plurality of past evaluation periods, a training data generation module is set to generate a training quadruple set required by a training deep reinforcement learning model based on the historical travel data and the emission basic data, a model training module is set to train a deep reinforcement learning model for generating a decision-making selected action for each travel task to be decided of the public institutions based on the training quadruple set, and a monitoring control module is set to use the deep reinforcement learning model to make decisions for the action selected by the travel task to be decided; the balance of staff experience and carbon emission is ensured, and the energy saving and emission reduction efficiency of public institutions is enhanced.
Description
Technical Field
The invention belongs to the technical field of carbon emission reduction of public institutions, and particularly relates to an emission reduction efficiency evaluation system suitable for green traveling of public institutions.
Background
Public institutions serve as government authorities, educational institutions, medical institutions, and the like, and are responsible for important social responsibilities and public service tasks. However, with the increase in motor vehicles and the acceleration of the urban process, public institutions face emissions reduction challenges in terms of vehicle use.
It is currently the case that police and private cars are widely used in public institutions, which results in a large amount of carbon emissions and environmental pollution. A utility vehicle is typically a vehicle that a public institution provides to employees for business trips, meetings, etc. and a private vehicle is the employee's own vehicle. Due to the need for travel tasks, these vehicles often travel frequently, producing significant carbon emissions.
However, existing travel decisions often lack emission reduction considerations. Decision makers tend to focus only on the completion and efficiency of tasks, and neglect the impact of vehicle use on the environment. This results in a large amount of carbon emissions, exacerbating environmental problems such as climate change and air pollution.
Therefore, the invention provides an emission reduction efficiency evaluation system suitable for green travel of public institutions.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides the emission reduction efficiency evaluation system suitable for the green travel of the public institution, which ensures the balance of staff experience and carbon emission, and enhances the energy conservation and emission reduction efficiency of the public institution.
To achieve the above objective, an embodiment according to a first aspect of the present invention provides an emission reduction efficiency evaluation system suitable for green travel of an institutional unit, including an emission information collection module, a travel data collection module, a training data generation module, a model training module, and an emission reduction decision module; wherein, each module is connected by a wired network mode;
wherein the emission information collection module collects emission base data of an institutional;
the emission base data includes emission efficiency data and emission vehicle data;
wherein the emission efficiency data includes an evaluation period for evaluating the carbon emission amount of the public institution and an allowable total carbon emission amount credit in each evaluation period;
wherein the emission vehicle data includes a public vehicle information set and a private vehicle information set that produce carbon emissions within the public institution;
each element in the public vehicle information set is public vehicle information, and the public vehicle information comprises a public vehicle number of a public service vehicle, unit fuel consumption corresponding to the public vehicle number and an emission factor corresponding to the public vehicle number;
each element in the private car information set is private car information, and the private car information comprises a private car number of a private car, a unit fuel consumption amount corresponding to the private car number and an emission factor corresponding to the private car number;
the emission information collection module sends the collected emission basic data to the training data generation module and the emission reduction decision module;
the travel data collection module collects a plurality of historical travel data of the public institution in the past T evaluation periods; wherein T is the preset acquisition period number;
the collection mode of the historical trip data is as follows:
collecting travel data of each vehicle in the public institution in time sequence in each evaluation period in the past; it can be understood that each trip data corresponds to a trip; the travel data comprise vehicle types, vehicle numbers, travel starting points, travel ending points, travel distances, travel duration, vehicle types, number of vehicles and public transportation substitution duration;
wherein the vehicle type is one of a public service vehicle or a private vehicle; the vehicle number is one of a public vehicle number or a private vehicle number corresponding to the vehicle type; the vehicle type is the purpose of using each vehicle to travel; the types of the vehicles comprise business trips, attendance, private vehicle public use and others; the public transportation substitution time length is the shortest time length needed by using public transportation travel from the starting point to the end point corresponding to each piece of travel data;
the driving distance is calculated by the following steps:
when the type of the vehicle in any piece of trip data is a official use vehicle, the driving distance comprises the distance of all the routes which are covered by the journey corresponding to the trip data; marking the vehicle type of the travel data as a business trip;
in any piece of travel data, when the type of the vehicle is private, the calculation mode of the travel distance is as follows:
a private car owner in each public institution submits a attendance start point and an attendance end point in advance, and when the attendance start point and the attendance end point in the trip data correspond to the attendance start point and the attendance end point respectively, the car type of the trip data is marked as attendance; generating a shortest route from the attendance start point and the attendance end point by using navigation software or a navigation system, and taking the distance of the shortest route as a driving distance;
if the travel starting point does not correspond to the attendance starting point or the travel ending point does not correspond to the attendance ending point in the travel data, marking the vehicle type of the travel data as private vehicles public or other vehicles;
marking travel data as private or otherwise:
generating a private vehicle public task by a public institution business trip management background, wherein the private vehicle public task comprises a private vehicle number, a task starting point and a task ending point of a private vehicle, judging whether a travel starting point and a travel ending point in travel data corresponding to the private vehicle number are respectively consistent with the private vehicle public task starting point and the task ending point after the private vehicle public task is generated, if so, marking the vehicle type of the travel data as the private vehicle public, taking the total travel distance of a travel corresponding to the travel data as a travel distance, and marking the private vehicle public task as executed; if the travel data are inconsistent, marking the vehicle type of the travel data as other vehicles;
the driving duration is the shortest duration required by the self-driving vehicle to travel from the travel starting point to the travel ending point generated by navigation software or a navigation system;
the travel data collection module sends the historical travel data to the training data generation module;
the training data generation module generates a training four-element set required by training the deep reinforcement learning model based on the historical trip data and the emission basic data;
each element in the training four-element group is a training four-element group; one vehicle type in one evaluation period in each training quadruple corresponds to travel data which are public for business trip, attendance or private vehicles;
the training quadruple is generated in the following way:
the number of the evaluation period is marked as t, and the number of the travel data in the t-th evaluation period is marked as ti;
the ti-th trip data for the t-th evaluation period:
taking the residual carbon emission limit corresponding to the ti-th trip data as an initial state;
the calculation mode of the residual carbon emission allowance corresponding to the initial state is as follows:
before calculating the ti-th travel data in the t-th evaluation period, all selected actions are the total Mti of carbon emission generated by the travel corresponding to the travel data without using an alternative scheme;
the calculation mode of the carbon emission C generated by the travel corresponding to each trip data is as follows: c=s x y; wherein s is the driving distance, x is the unit fuel consumption corresponding to the vehicle number, and y is the emission factor corresponding to the vehicle number;
marking the allowable total carbon emission credit as Mmax for each evaluation period; the remaining carbon emission credit corresponding to the initial state is Mmax-Mti;
for the ti-th trip data, randomly selecting a decision action with 50% probability as the selected action; the decision action is either selecting or not selecting an alternative; the alternative scheme is selected or not selected to be used, namely a public transportation alternative scheme with the shortest travel time is selected or not selected to be used;
for the ti-th trip data, calculating a reward value Q brought by the selected action;
the calculation mode of the reward value Q is as follows:
for the ti trip data, calculating the carbon emissions Cti generated after the selected action is an unselected alternative; marking the number of vehicles in the ti-th travel data as Nti, and marking the public transportation substitution duration in the ti-th travel data as Kti; marking the running duration in the ti-th travel data as Jti;
if the residual carbon emission allowance is smaller than 0 after the selection action, the rewarding value Q is-Qmax; wherein QMax is a preset prize value threshold greater than 0;
otherwise, if the vehicle type of the travel data is a business trip, the calculation formula of the rewarding value Q is as follows:
Q=w*（a1*Nti*(Kti-Jti) -Cti）+(1-w)*b*( Mmax-Mti)；
if the type of the vehicle used for the travel data is attendance, the calculation formula of the rewarding value Q is as follows:
Q=w*（a2*Nti*(Kti-Jti)- Cti）+(1-w)*b*( Mmax-Mti)；
if the vehicle type of the travel data is public, the calculation formula of the rewarding value Q is as follows:
Q=w*（a3*Nti*(Kti-Jti)-Cti）+(1-w)*b*( Mmax-Mti)；
wherein a1, a2, a3 and b are respectively preset proportionality coefficients; w is one of 0 or 1, w=1 when the selected action is an alternative to be used; otherwise, w=0;
taking the residual carbon emission allowance after the selecting action of the ti travel data as the next state;
the training quadruple comprises an initial state, a selected action, a reward value Q and a next state which correspond to each piece of trip data;
the training data generation module sends the training four-element group set to the model training module;
the model training module trains a deep reinforcement learning model for generating a decision-making selection action for each trip task to be decided of the public institution based on the training four-element set;
the way to train the deep reinforcement learning model that generates the decision-selected actions for each piece of travel data of the public institution is:
the training quadruple set is used as input of a deep reinforcement learning model, the deep reinforcement learning model carries out training by randomly extracting a plurality of quadruples from the training quadruple set, and the strategy of obtaining the maximum rewarding value Q by deciding to select or not select to use an alternative scheme under different initial states is learned; the deep reinforcement learning model is a deep Q network model;
the model training module sends the trained deep reinforcement learning model to the monitoring control module;
the monitoring control module uses a deep reinforcement learning model to make decisions on actions selected by the trip task to be decided;
the mode of deciding the action to be decided for selecting the trip data is as follows:
the business trip management background receives the travel task to be decided of each public service vehicle and each private vehicle in real time, and calculates the residual carbon emission limit in the current evaluation period when the travel task to be decided is newly generated;
and taking the residual carbon emission amount as the input of the deep reinforcement learning model, and outputting the decision of the selected action by the deep reinforcement learning model.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, the system and the device, the emission basic data of the public institution are collected in advance, the travel vehicles in the public institution are divided, a plurality of historical travel data of the public institution in a plurality of past evaluation periods are collected, travel data of different vehicles in the historical data are classified according to official vehicles and private vehicles, different vehicle types are further divided, corresponding travel distances are calculated according to different vehicle types, a public transportation alternative scheme is distributed for each travel data, a training quaternary set required by a training depth reinforcement learning model is generated based on the historical travel data and the emission basic data, in the quaternary set, the calculation mode of a reward value Q is different according to different vehicle types, so that the flexibility of action decision is improved, a depth reinforcement learning model for generating decision-making selected actions for each travel task to be decided of the public institution is trained based on the training quaternary set, and finally the decision-making task to be decided is performed by using the depth reinforcement learning model; therefore, whether travel tasks of each public service vehicle and each private vehicle adopt an alternative scheme or not is intelligently decided in the evaluation period of each carbon emission, balance of employee experience and carbon emission is ensured, and energy conservation and emission reduction efficiency of public institutions is enhanced.
Drawings
Fig. 1 is a diagram of a module connection relationship of an emission reduction efficiency evaluation system suitable for green travel of public institutions.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in FIG. 1, the emission reduction efficiency evaluation system suitable for green travel of public institutions is used in a business trip management background and comprises an emission information collection module, a travel data collection module, a training data generation module, a model training module and an emission reduction decision module; wherein, each module is connected by a wired network mode;
wherein the emission information collection module is mainly used for collecting emission basic data of public institutions;
in a preferred embodiment, the emissions base data includes emissions efficiency data and emissions vehicle data;
wherein the emission efficiency data includes an evaluation period for evaluating the carbon emission amount of the public institution and an allowable total carbon emission amount credit in each evaluation period;
wherein the emission vehicle data includes a public vehicle information set and a private vehicle information set that produce carbon emissions within the public institution;
each element in the public vehicle information set is public vehicle information, and the public vehicle information comprises a public vehicle number of a public service vehicle, unit fuel consumption corresponding to the public vehicle number and an emission factor corresponding to the public vehicle number; it is understood that the unit fuel consumption refers to a fuel consumption per kilometer, which varies in value according to different vehicle types; the emission factor is determined according to the type of fuel used, for example, carbon dioxide emissions of about 2.3 to 2.7 kg carbon dioxide per liter of gasoline or diesel are produced; the unit fuel consumption and the emission factor are determined according to each service vehicle;
each element in the private car information set is private car information, and the private car information comprises a private car number of a private car, a unit fuel consumption amount corresponding to the private car number and an emission factor corresponding to the private car number;
the emission information collection module sends the collected emission basic data to the training data generation module and the emission reduction decision module;
the travel data collection module is mainly used for collecting a plurality of historical travel data of public institutions in the past T evaluation periods; wherein T is the preset acquisition period number;
the collection mode of the historical trip data is as follows:
collecting travel data of each vehicle in the public institution in time sequence in each evaluation period in the past; it can be understood that each trip data corresponds to a trip; the travel data comprise vehicle types, vehicle numbers, travel starting points, travel ending points, travel distances, travel duration, vehicle types, number of vehicles and public transportation substitution duration;
wherein the vehicle type is one of a public service vehicle or a private vehicle; the vehicle number is one of a public vehicle number or a private vehicle number corresponding to the vehicle type; the vehicle type is the purpose of using each vehicle to travel; specifically, the types of vehicles include business trips, attendance, private use and others; the public transportation substitution time length is the shortest time length needed by using public transportation travel from the starting point to the end point corresponding to each piece of travel data; it can be understood that the existing navigation software has the functions of inputting a starting point and a destination point and evaluating various travel modes and the shortest duration of travel routes, and the invention does not describe how to obtain the public transportation substitution duration again;
the travel distance is calculated by:
when the type of the vehicle in any piece of trip data is a official use vehicle, the driving distance comprises the distance of all the routes which are covered by the journey corresponding to the trip data; marking the vehicle type of the travel data as a business trip; it can be understood that the distance of all the routes traveled by the journey can be obtained by navigation software or a navigation system, and will not be described in detail herein;
in any piece of travel data, when the type of the vehicle is private, the calculation mode of the travel distance is as follows:
a private car owner in each public institution submits a attendance start point and an attendance end point in advance, and when the attendance start point and the attendance end point in the trip data correspond to the attendance start point and the attendance end point respectively, the car type of the trip data is marked as attendance; generating a shortest route from the attendance start point and the attendance end point by using navigation software or a navigation system, and taking the distance of the shortest route as a driving distance; it can be understood that the distance of the shortest route is taken as the driving distance, only the attendance route of the private car in each journey is calculated, and excessive private routes added into the private car are avoided, so that the experience of staff in a public institution is reduced;
the driving duration is the shortest duration required by the self-driving vehicle to travel from the travel starting point to the travel ending point generated by navigation software or a navigation system;
further, if the trip starting point does not correspond to the attendance starting point or the trip ending point does not correspond to the attendance ending point in the trip data, marking the trip data as private car public or other trip data;
in a preferred embodiment, marking travel data as private or otherwise:
generating a private vehicle public task by a public institution business trip management background, wherein the private vehicle public task comprises a private vehicle number, a task starting point and a task ending point of a private vehicle, judging whether a travel starting point and a travel ending point in travel data corresponding to the private vehicle number are respectively consistent with the private vehicle public task starting point and the task ending point after the private vehicle public task is generated, if so, marking the vehicle type of the travel data as the private vehicle public, taking the total travel distance of a travel corresponding to the travel data as a travel distance, and marking the private vehicle public task as executed; if the travel data are inconsistent, marking the vehicle type of the travel data as other vehicles; it can be understood that the private vehicle public task is of a common type of public institutions, and the private vehicle owners of the private vehicles need to pay for the fee, and corresponding business trip management background exists in the public institutions, so that the invention is not repeated here; further, when the private car public task is marked as executed, the type of the vehicle of the travel data, of which the subsequent travel starting point and travel destination are respectively consistent with the private car public task starting point and the task destination, is not marked as private car public any more;
the driving duration is the shortest duration required by the self-driving vehicle to travel from the travel starting point to the travel ending point generated by navigation software or a navigation system;
the travel data collection module sends the historical travel data to the training data generation module;
the training data generation module is mainly used for generating a training four-element set required by training the deep reinforcement learning model based on the historical trip data and the emission basic data;
each element in the training four-element group is a training four-element group; one vehicle type in one evaluation period in each training quadruple corresponds to travel data which are public for business trip, attendance or private vehicles;
the training quadruple is generated in the following way:
the number of the evaluation period is marked as t, and the number of the travel data in the t-th evaluation period is marked as ti;
the ti-th trip data for the t-th evaluation period:
taking the residual carbon emission limit corresponding to the ti-th trip data as an initial state;
preferably, the remaining carbon emission allowance corresponding to the initial state is calculated in the following manner:
before calculating the ti-th travel data in the t-th evaluation period, all selected actions are the total Mti of carbon emission generated by the travel corresponding to the travel data without using an alternative scheme;
the calculation mode of the carbon emission C generated by the travel corresponding to each trip data is as follows: c=s x y; wherein s is the driving distance, x is the unit fuel consumption corresponding to the vehicle number, and y is the emission factor corresponding to the vehicle number;
marking the allowable total carbon emission credit as Mmax for each evaluation period; the remaining carbon emission credit corresponding to the initial state is Mmax-Mti;
for the ti-th trip data, randomly selecting a decision action with 50% probability as the selected action; the decision action is either selecting or not selecting an alternative; the alternative scheme is selected or not selected to be used, namely a public transportation alternative scheme with the shortest travel time is selected or not selected to be used; it should be noted that, because each trip data in the history data is a travel that has occurred, in the process of generating the training quadruple, the selected action needs to be adjusted;
for the ti-th trip data, calculating a reward value Q brought by the selected action;
the calculation mode of the reward value Q is as follows:
for the ti trip data, calculating the carbon emissions Cti generated after the selected action is an unselected alternative; marking the number of vehicles in the ti-th travel data as Nti, and marking the public transportation substitution duration in the ti-th travel data as Kti; marking the running duration in the ti-th travel data as Jti;
if the residual carbon emission allowance is smaller than 0 after the selection action, the rewarding value Q is-Qmax; wherein QMax is a preset prize value threshold greater than 0;
otherwise, if the vehicle type of the travel data is a business trip, the calculation formula of the rewarding value Q is as follows:
Q=w*（a1*Nti*(Kti-Jti) -Cti）+(1-w)*b*( Mmax-Mti)；
if the type of the vehicle used for the travel data is attendance, the calculation formula of the rewarding value Q is as follows:
Q=w*（a2*Nti*(Kti-Jti)- Cti）+(1-w)*b*( Mmax-Mti)；
if the vehicle type of the travel data is public, the calculation formula of the rewarding value Q is as follows:
Q=w*（a3*Nti*(Kti-Jti)-Cti）+(1-w)*b*( Mmax-Mti)；
wherein a1, a2, a3 and b are respectively preset proportionality coefficients; w is one of 0 or 1, w=1 when the selected action is an alternative to be used; otherwise, w=0; it can be appreciated that when the number of people on the trip is greater or the time gap between the alternative and the use of the public or private vehicles is greater, the alternative is more prone to be not used so as to protect the trip mood of staff; when the remaining carbon emission amount credit is smaller or the carbon emission amount corresponding to the ti-th trip data is larger, an alternative scheme is more prone to be used so as to improve the emission reduction efficiency;
taking the residual carbon emission allowance after the selecting action of the ti travel data as the next state;
the training quadruple comprises an initial state, a selected action, a reward value Q and a next state which correspond to each piece of trip data;
the training data generation module sends the training four-element group set to the model training module;
the model training module is mainly used for training a deep reinforcement learning model for generating decision selection actions for each trip task to be decided of the public institution based on the training four-element set; the travel task to be decided is a travel task which is needed to be carried out on business trips, attendance or private vehicles, but is not executed yet; the business trip management background decides whether the travel task to be decided is carried out by using a public transportation alternative scheme so as to improve the emission reduction efficiency; the travel task to be decided comprises a vehicle type, a vehicle number, a travel starting point, a travel ending point, a travel distance, a vehicle type, a vehicle number and a public transportation substitution duration; the travel distance of the travel task to be decided is the distance from the travel starting point to the travel ending point, which is generated by the navigation software;
the way to train the deep reinforcement learning model that generates the decision-selected actions for each piece of travel data of the public institution is:
the training quadruple set is used as input of a deep reinforcement learning model, the deep reinforcement learning model carries out training by randomly extracting a plurality of quadruples from the training quadruple set, and the strategy of obtaining the maximum rewarding value Q by deciding to select or not select to use an alternative scheme under different initial states is learned; the deep reinforcement learning model is a deep Q network model;
the model training module sends the trained deep reinforcement learning model to the monitoring control module;
the monitoring control module is mainly used for deciding actions selected by the travel task to be decided by using a deep reinforcement learning model;
preferably, the way of deciding the action selected by the trip data to be decided is as follows:
the business trip management background receives the travel task to be decided of each public service vehicle and each private vehicle in real time, and calculates the residual carbon emission limit in the current evaluation period when the travel task to be decided is newly generated;
and taking the residual carbon emission amount as the input of the deep reinforcement learning model, and outputting the decision of the selected action by the deep reinforcement learning model.
The above preset parameters or preset thresholds are set by those skilled in the art according to actual conditions or are obtained by mass data simulation.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (9)
1. The emission reduction efficiency evaluation system suitable for the green travel of the public institution is characterized by comprising an emission information collection module, a travel data collection module, a training data generation module, a model training module and an emission reduction decision module; wherein, each module is connected by a wired network mode;
the emission information collection module is used for collecting emission basic data of the public institution and sending the collected emission basic data to the training data generation module and the emission reduction decision module;
the travel data collection module is used for collecting a plurality of historical travel data of the public institution in the past T evaluation periods and sending the historical travel data to the training data generation module; wherein T is the preset acquisition period number;
the training data generation module is used for generating a training four-element set required by training the deep reinforcement learning model based on the historical trip data and the emission basic data, and transmitting the training four-element set to the model training module;
the model training module is used for training a deep reinforcement learning model for generating a decision-making selected action for each trip task to be decided of the public institution based on the training four-element set, and transmitting the trained deep reinforcement learning model to the monitoring control module;
and the monitoring control module is used for deciding the action selected by the travel task to be decided by using the deep reinforcement learning model.
2. An emission reduction efficiency evaluation system for institutional green travel as claimed in claim 1, wherein the emission base data includes emission efficiency data and emission vehicle data;
wherein the emission efficiency data includes an evaluation period for evaluating the carbon emission amount of the public institution and an allowable total carbon emission amount credit in each evaluation period;
wherein the emission vehicle data includes a public vehicle information set and a private vehicle information set that generate carbon emissions within the public institution.
3. The emission reduction efficiency evaluation system suitable for green travel of public institutions according to claim 2, wherein each element in the public vehicle information set is public vehicle information, and the public vehicle information comprises a public vehicle number of a public service vehicle, unit fuel consumption corresponding to the public vehicle number and emission factors corresponding to the public vehicle number;
each element in the private car information set is private car information, and the private car information comprises a private car number of a private car, unit fuel consumption corresponding to the private car number and an emission factor corresponding to the private car number.
4. The emission reduction efficiency evaluation system suitable for green travel of public institutions according to claim 3, wherein the collection mode of the historical travel data is as follows:
collecting travel data of each vehicle in the public institution in time sequence in each evaluation period in the past; it can be understood that each trip data corresponds to a trip; the travel data comprise vehicle types, vehicle numbers, travel starting points, travel ending points, travel distances, travel duration, vehicle types, number of vehicles and public transportation substitution duration;
the vehicle type is one of a public service vehicle or a private vehicle; the vehicle number is one of a public vehicle number or a private vehicle number corresponding to the vehicle type; the vehicle type is the purpose of using each vehicle to travel; the types of the vehicles comprise business trips, attendance, private vehicle public use and others; the public transportation substitution time length is the shortest time length needed by using public transportation travel from the starting point to the end point corresponding to each piece of travel data;
the driving duration is the shortest duration required by the self-driving vehicle to travel from the travel starting point to the travel ending point generated by navigation software or a navigation system.
5. The emission reduction efficiency evaluation system for green travel of public institutions according to claim 4, wherein the travel distance is calculated by:
when the type of the vehicle in any piece of trip data is a official use vehicle, the driving distance comprises the distance of all the routes which are covered by the journey corresponding to the trip data; marking the vehicle type of the travel data as a business trip;
in any piece of travel data, when the type of the vehicle is private, the calculation mode of the travel distance is as follows:
a private car owner in each public institution submits a attendance start point and an attendance end point in advance, and when the attendance start point and the attendance end point in the trip data correspond to the attendance start point and the attendance end point respectively, the car type of the trip data is marked as attendance; generating a shortest route from the attendance start point and the attendance end point by using navigation software or a navigation system, and taking the distance of the shortest route as a driving distance;
if the travel starting point does not correspond to the attendance starting point or the travel ending point does not correspond to the attendance ending point in the travel data, marking the vehicle type of the travel data as private vehicles public or other vehicles;
marking travel data as private or otherwise:
generating a private vehicle public task by a public institution business trip management background, wherein the private vehicle public task comprises a private vehicle number, a task starting point and a task ending point of a private vehicle, judging whether a travel starting point and a travel ending point in travel data corresponding to the private vehicle number are respectively consistent with the private vehicle public task starting point and the task ending point after the private vehicle public task is generated, if so, marking the vehicle type of the travel data as the private vehicle public, taking the total travel distance of a travel corresponding to the travel data as a travel distance, and marking the private vehicle public task as executed; and if the travel data are inconsistent, marking the vehicle type of the travel data as other vehicles.
6. The emission reduction efficiency evaluation system suitable for green travel of an institutional entity of claim 5, wherein each element in the training four-tuple set is a training four-tuple; one vehicle type in one evaluation period in each training quadruple corresponds to travel data which are public for business trip, attendance or private vehicles;
the training quadruple is generated in the following way:
the number of the evaluation period is marked as t, and the number of the travel data in the t-th evaluation period is marked as ti;
the ti-th trip data for the t-th evaluation period:
taking the residual carbon emission limit corresponding to the ti-th trip data as an initial state;
for the ti-th trip data, randomly selecting a decision action with 50% probability as the selected action; the decision action is either selecting or not selecting an alternative; the alternative scheme is a public transportation alternative scheme with the shortest travel time length selected or not selected;
for the ti-th trip data, calculating a reward value Q brought by the selected action;
taking the residual carbon emission allowance after the selecting action of the ti travel data as the next state;
the training quadruple comprises an initial state, a selected action, a reward value Q and a next state which correspond to each piece of trip data.
7. The emission reduction efficiency evaluation system suitable for green travel of public institutions according to claim 6, wherein the reward value Q is calculated by the following manner:
for the ti trip data, calculating the carbon emissions Cti generated after the selected action is an unselected alternative; marking the number of vehicles in the ti-th travel data as Nti, and marking the public transportation substitution duration in the ti-th travel data as Kti; marking the running duration in the ti-th travel data as Jti;
if the residual carbon emission allowance is smaller than 0 after the selection action, the rewarding value Q is-Qmax; wherein QMax is a preset prize value threshold greater than 0;
otherwise, if the vehicle type of the travel data is a business trip, the calculation formula of the rewarding value Q is as follows:
Q=w*（a1*Nti*(Kti-Jti) -Cti）+(1-w)*b*( Mmax-Mti)；
if the type of the vehicle used for the travel data is attendance, the calculation formula of the rewarding value Q is as follows:
Q=w*（a2*Nti*(Kti-Jti)- Cti）+(1-w)*b*( Mmax-Mti)；
if the vehicle type of the travel data is public, the calculation formula of the rewarding value Q is as follows:
Q=w*（a3*Nti*(Kti-Jti)-Cti）+(1-w)*b*( Mmax-Mti)；
wherein a1, a2, a3 and b are respectively preset proportionality coefficients; w is one of 0 or 1, w=1 when the selected action is an alternative to be used; otherwise, w=0.
8. The emission reduction efficiency evaluation system for green travel of an institutional entity of claim 7 wherein training a deep reinforcement learning model to generate decision-selected actions for each piece of travel data of the institutional entity is:
the training quadruple set is used as input of a deep reinforcement learning model, the deep reinforcement learning model carries out training by randomly extracting a plurality of quadruples from the training quadruple set, and the strategy of obtaining the maximum rewarding value Q by deciding to select or not select to use an alternative scheme under different initial states is learned; the deep reinforcement learning model is a deep Q network model.
9. The emission reduction efficiency evaluation system for green travel of public institutions according to claim 8, wherein the decision making mode of the action to be decided on the travel data selection is as follows:
the business trip management background receives the travel task to be decided of each public service vehicle and each private vehicle in real time, and calculates the residual carbon emission limit in the current evaluation period when the travel task to be decided is newly generated;
and taking the residual carbon emission amount as the input of the deep reinforcement learning model, and outputting the decision of the selected action by the deep reinforcement learning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311050021.XA CN116757559B (en) | 2023-08-21 | 2023-08-21 | Emission reduction efficiency evaluation system suitable for green trip of public institution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311050021.XA CN116757559B (en) | 2023-08-21 | 2023-08-21 | Emission reduction efficiency evaluation system suitable for green trip of public institution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116757559A true CN116757559A (en) | 2023-09-15 |
| CN116757559B CN116757559B (en) | 2023-10-27 |
Family
ID=87953717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311050021.XA Active CN116757559B (en) | 2023-08-21 | 2023-08-21 | Emission reduction efficiency evaluation system suitable for green trip of public institution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116757559B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114819305A (en) * | 2022-04-13 | 2022-07-29 | 山东高速云南发展有限公司 | Path planning method based on carbon emission measurement scale |
| CN115619037A (en) * | 2022-11-07 | 2023-01-17 | 河海大学 | Multi-type combined transport multi-target decision research method for quantifying carbon emission influence |
| CN115691136A (en) * | 2022-11-01 | 2023-02-03 | 西安航空学院 | Urban low-carbon passenger traffic structure evaluation system based on data analysis |
-
2023
- 2023-08-21 CN CN202311050021.XA patent/CN116757559B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114819305A (en) * | 2022-04-13 | 2022-07-29 | 山东高速云南发展有限公司 | Path planning method based on carbon emission measurement scale |
| CN115691136A (en) * | 2022-11-01 | 2023-02-03 | 西安航空学院 | Urban low-carbon passenger traffic structure evaluation system based on data analysis |
| CN115619037A (en) * | 2022-11-07 | 2023-01-17 | 河海大学 | Multi-type combined transport multi-target decision research method for quantifying carbon emission influence |
Non-Patent Citations (1)
| Title |
|---|
| MAJID ZAHIRI: "可持续城市交通系统车辆小型化与共享出行政策", CNKI博士学位论文全文库 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116757559B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8682572B2 (en) | Method and device for evaluating vehicle's fuel consumption efficiency | |
| Boriboonsomsin et al. | Eco-driving: pilot evaluation of driving behavior changes among us drivers | |
| Lee et al. | Synthesis of real-world driving cycles and their use for estimating PHEV energy consumption and charging opportunities: Case study for Midwest/US | |
| US20150310467A1 (en) | Shared vehicle systems and methods | |
| CN109784560A (en) | A kind of electric car course continuation mileage evaluation method and estimating system | |
| CN109145982A (en) | The personal identification method and device of driver, storage medium, terminal | |
| CN110118567A (en) | Trip mode recommended method and device | |
| CN109359169B (en) | Shared bicycle reverse behavior real-time identification method based on probability map model | |
| CN110400462B (en) | Track traffic passenger flow monitoring and early warning method and system based on fuzzy theory | |
| CN104680348A (en) | System for evaluating performance of drivers of logistics vehicles | |
| Lopez-Ruiz et al. | Quantifying the effects of sustainable urban mobility plans | |
| Ojeda et al. | Fuel consumption prediction for heavy-duty vehicles using digital maps | |
| CN102063645A (en) | Public traffic operation service quality evaluating system | |
| CN116757559B (en) | Emission reduction efficiency evaluation system suitable for green trip of public institution | |
| CN110019376B (en) | Driving comparison method for electric vehicle rental system | |
| CN204423430U (en) | A kind of logistics vehicles driver performance appraisal system | |
| CN104616066A (en) | A method and system for collecting data of usage amount of travelling vehicles | |
| Rahadian et al. | Enabling Clean and Smart Mobility in Indonesian Road Infrastructure to Achieve Sustainable Transport | |
| Markel et al. | Transportation electrification load development for a renewable future analysis | |
| Chakrapani et al. | Sustainable Smart Energy Systems and Energy Preservation Strategies in Intelligent Transportation Sectors | |
| US20240062310A1 (en) | Automated method and system for determining an expected damage of an electrically powered vehicle | |
| Sabet et al. | Exploring sustainable pathways for urban traffic decarbonization: vehicle technologies, management strategies, and driving behaviour | |
| CN113554770B (en) | Method, device and equipment for acquiring staggered account information and readable storage medium | |
| Openshaw et al. | Investigation of Commercial Viability and Public Perception of Electrified Roadways with Dynamic Wireless Charging | |
| Monghasemi et al. | Time-Environmental Impacts Tradeoff Analysis for Businesses Commuters |
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 |