Internet of Things service management system based on regional digital economy

Abstract

The invention discloses an Internet of Things service management system based on regional digital economy, which relates to the technical field of Internet of Things and comprises a background charging planning module, a plurality of charging stations arranged in a high-speed service area, and a positioning module and a navigation module arranged on an electric vehicle. According to the invention, through the cooperative arrangement of the parking lock and the background charging planning module, the car owner can be prevented from scrambling for the charging position in the peak period, so that the background charging planning module can uniformly and orderly plan charging.

Description

Technical Field

[0001] The present invention relates to the technical field of the Internet of Things, and more specifically, to an Internet of Things service management system based on regional digital economy. Background Art

[0002] Digital economy is the main economic form after agricultural economy and industrial economy. It is a new economic form that takes data resources as key elements, modern information networks as the main carriers, the integrated application of information and communication technologies and the full-factor digital transformation as important driving forces, and promotes the more unified fairness and efficiency. The digital economy develops rapidly, has a wide radiation range, and has a deep impact. At the technical level, it includes emerging technologies such as big data, cloud computing, the Internet of Things, blockchain, artificial intelligence, and 5G communication. At the application level, "new retail", "new manufacturing", etc. are all its typical representatives.

[0003] With the increase in the number of electric vehicles, the charging piles in highway service areas or charging stations have become scarce resources. Due to the charging time of electric vehicles, it is easy to cause the accumulation of electric vehicles at the charging station, with too long waiting time for charging. Moreover, during the process of vehicle owners waiting for charging, it is easy to cause chaos at the scene, leading to the occurrence of safety accidents.

[0004] In the Chinese invention patent with the application number CN202211240228.9, there is disclosed an Internet of Things service management system based on regional digital economy. Although it can predict the time when the target electric vehicle to be charged arrives at each public charging station according to its planned driving route and road conditions, and then arrange the target electric vehicle according to the change in the number of charging people at each public charging station during this period to ensure that the target electric vehicle can minimize the queuing time within the cruising range, however, this Internet of Things service management system based on regional digital economy still cannot avoid the situation where the charging piles in front of the charging station are occupied by the electric vehicles planned in advance.

[0005] Even if there may be technologies in the prior art that can reserve charging piles, they still cannot avoid non-charging electric vehicles staying in the charging positions of the charging piles, resulting in the inability of the electric vehicles planned to reach this charging pile to charge.

[0006] Therefore, it is very necessary to propose an Internet of Things service management system based on regional digital economy to solve the above problems. Summary of the Invention

[0007] The purpose of the present invention is to solve the problems mentioned in the background art.

[0008] The present invention specifically adopts the following technical solutions to achieve the above purpose:

[0009] The Internet of Things service management system based on regional digital economy includes a background charging planning module, multiple charging stations set in highway service areas, and a positioning module and a navigation module set on electric vehicles;

[0010] Each charging station is provided with multiple rows of electric vehicle charging piles. Each charging pile is provided with two charging spaces, which are symmetrically arranged on the front and back sides of the charging pile, and each space is provided with a ground lock. When the electric vehicle finishes charging, the ground lock automatically opens. When the electric vehicle leaves the charging pile after charging, the ground lock automatically closes;

[0011] The background charging planning module is used to plan and allocate vehicle charging piles according to the electric vehicle power and location information, and pre-store the license plate number of the electric vehicle in the ground lock of the charging pile;

[0012] When the positioning module monitors that the electric vehicle passes through the charging station planned by the background charging planning module, the background charging planning module cancels the license plate number of the electric vehicle pre-stored in the ground lock of the charging pile;

[0013] A license plate recognition module is set in the ground lock, which is used to recognize the electric vehicle planned by the background charging planning module to this location. When the recognition is correct, the ground lock automatically opens; if the recognition fails, the owner is prompted to perform manual verification through the navigation module on the electric vehicle;

[0014] When the electric vehicle charging at the charging pile suddenly disconnects and stops charging and the electric vehicle drives out of the charging pile, the background charging planning module plans the electric vehicle that was originally navigated to this charging station and is the closest to this charging station to enter this charging station, and plans the navigation module to navigate the vehicle into the charging pile where charging suddenly disconnects and stops;

[0015] A redundant communication mechanism is provided between the ground lock and the background charging planning module. When the network is interrupted, the ground lock completes the verification through the license plate information cached locally.

[0016] Further, the background charging planning module analyzes whether it is the peak traffic flow period based on real-time traffic data analysis. During the peak traffic flow period, the charging station is only for pure electric vehicles, and this charging pile is not pushed to hybrid electric vehicles;

[0017] The peak traffic flow period is judged by real-time monitoring of the number of vehicles entering the service area or the charging pile occupancy rate. When the number of vehicles entering exceeds the preset threshold or the charging pile occupancy rate ≥ 80%, the system automatically enters the peak period mode.

[0018] Further, the charging pile is single-gun. When all the charging piles are occupied, the background charging planning module plans the electric vehicle to the space on the opposite side of the charging pile with the most remaining charge.

[0019] Furthermore, a dynamic power distribution module, a priority adjustment module, and a power limit optimization module are provided in the background charging planning module;

[0020] The dynamic power distribution module dynamically adjusts the power distribution for each charging pile according to the battery capacity, remaining power, charging demand of the currently charging vehicle, and the overall power load of the charging station:

[0021]

[0022] Where: P i The allocated power of the i-th charging pile, G i The battery capacity of the vehicle connected to the i-th charging pile, S i The percentage of remaining power of the vehicle connected to the i-th charging pile, P total The total power of the charging station, and n is the number of vehicles currently charging in the charging station;

[0023] For vehicles with low remaining power, the priority adjustment module automatically increases their charging priority:

[0024]

[0025] Where: Priority i The priority of the i-th charging pile, G max The maximum battery capacity of all vehicles in the current charging station;

[0026] When the total power of the charging station is close to the upper limit, the power limit optimization module automatically reduces the charging power of low-priority vehicles to ensure that high-priority vehicles can obtain sufficient charging power:

[0027]

[0028] Where: P' i Is the power of the adjusted i-th charging pile, P max The maximum power limit of the charging station, P used The total power already used in the current charging station.

[0029] Furthermore, the background charging planning module is used to predict the charging demand of each service area in real time and provide input data for the scheduling algorithm:

[0030] Q(t, s) = σ(Conv3D(W conv *[H(t - Δt), T(t), W(s)] + b conv ));

[0031] Where: Q(t, s) is the predicted charging demand of service area s at time t, unit: vehicles per hour, H(t - Δt) is the time series data of the charging amount in the historical Δt period, with dimension K×1, where Δt is the size of the time window, T(t) is the real-time traffic flow matrix, with dimension N×N, representing the traffic conditions of the road network around the service area, W conv is the three-dimensional convolution kernel weight, used for spatio-temporal feature extraction, b conv is the bias term of the convolutional layer, σ is the LeakyReLU activation function, defined as:

[0032]

[0033] W(s) is the weather influence factor, calculated by cubic polynomial fitting of humidity h and temperature τ:

[0034] W(s) = a0 + a1h + a2τ + a3h 2 + a4τ 2 + a5h 3 + a6τ 3 .

[0035] Furthermore, the scheduling algorithm is based on the predicted charging demand Q(t, s), and the dynamic priority scheduling algorithm optimizes the allocation of charging piles:

[0036]

[0037] The constraint conditions are:

[0038]

[0039] Where: U is the total number of electric vehicles to be scheduled, V is the total number of charging piles in the charging station, W is the number of supply-demand balance constraints, x uv is the binary decision variable, indicating whether vehicle u is assigned to charging pile v, x uv = 1 indicates assignment, x uv = 0 indicates non-assignment, c uv is the comprehensive cost of vehicle u being assigned to charging pile v, and the calculation formula is:

[0040]

[0041] Where: d uv is the distance from vehicle u to charging pile v, d max is the maximum allowable distance in the system, E u is the remaining battery power of the vehicle, E max is the maximum capacity of the vehicle's battery, t wait is the estimated waiting time of vehicle u, t max is the maximum allowable waiting time in the system, α, β, γ are weight coefficients, satisfying α + β + γ = 1, Cv is the maximum capacity of charging pile v (i.e., the number of vehicles that can be served simultaneously), R min is the minimum power requirement for vehicle charging, Q(t, s) is the predicted charging demand of service area s at time t provided by the charging demand dynamic prediction model, y w is the slack variable of the supply-demand balance constraint, representing the deviation between the actual allocation and the demand prediction, ∈ is the maximum allowable deviation of the supply-demand balance, λ is the penalty factor used to control the impact of the supply-demand balance deviation, and the calculation formula is:

[0042]

[0043] In the formula: C total is the total capacity of all charging piles in the system.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] 1. In the present invention, through the coordinated setting between the ground lock and the background charging planning module, it is possible to avoid the scramble of vehicle owners for charging positions during peak periods, thereby enabling the unified and orderly planning of charging by the background charging planning module.

[0046] 2. In the present invention, through the dynamic power distribution algorithm, the system can optimize the power according to the real-time needs of vehicles, improve the charging efficiency, reduce the power waste of the charging station, and meet the charging needs of different vehicles.

[0047] 3. In the present invention, through the three-dimensional convolution operation that fuses multi-dimensional spatio-temporal features, the prediction accuracy is improved compared with the traditional ARIMA model.

[0048] 4. In the present invention, the supply-demand balance slack variable y is introduced for the first time w , so that the scheduling scheme remains feasible in case of emergencies, and the measured scheduling success rate is improved.

[0049] 5. In the present invention, the integration of prediction and scheduling closely combines the charging demand prediction and the scheduling algorithm, realizes the dynamic optimization prediction model to provide high-precision input, and the scheduling algorithm ensures the efficient use of resources. Detailed implementation manners

[0050] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0051] An IoT service management system based on the regional digital economy, including a backend charging planning module, multiple charging stations set up in high-speed service areas, and a positioning module and a navigation module set up on electric vehicles;

[0052] In order to realize the planning of the background charging planning module, the requirements for charging stations are that each charging station is equipped with multiple rows of electric vehicle charging piles, each charging pile is equipped with two charging parking spaces, and they are symmetrically arranged on the front and back sides of the charging pile, and each parking space is equipped with a ground lock. When the electric vehicle is fully charged, the ground lock automatically opens. When the electric vehicle leaves the charging pile after charging, the ground lock automatically closes to prevent other vehicles from cutting in after the vehicle is fully charged and leaves. If forced to enter, the ground lock will lock the charging pile;

[0053] In addition, some reminder mechanisms can be set up to remind vehicles not to enter the parking space to avoid subsequent accidents.

[0054] The backend charging planning module is used to plan and allocate the electric vehicle to the vehicle charging pile according to the electric power and location information of the electric vehicle, and pre-store the license plate number of the electric vehicle in the ground lock of the charging pile, so as to achieve the predetermined effect and make the charging planning for the charging vehicle orderly;

[0055] Since vehicles cannot turn around on the highway, when the positioning module detects that the electric vehicle passes through the charging station planned by the background charging planning module, the background charging planning module cancels the license plate number of the electric vehicle pre-stored in the ground lock of the charging pile;

[0056] The ground lock is equipped with a license plate recognition module, which is used to identify the electric vehicle that is planned to the location by the background charging planning module. When the recognition is correct, the ground lock is automatically opened; if the recognition fails, the navigation module on the electric vehicle prompts the owner to perform manual verification;

[0057] When an electric vehicle that is charging at a charging pile suddenly disconnects and stops charging and the electric vehicle drives out of the charging pile, the background charging planning module plans the electric vehicle that was originally navigated to the charging station and is closest to the charging station to enter the charging station, and the planning navigation module navigates the vehicle to the charging pile that suddenly disconnects and stops charging;

[0058] A redundant communication mechanism is set up between the ground lock and the background charging planning module. When the network is interrupted, the ground lock completes the verification through the locally cached license plate information.

[0059] Specifically, the background charging planning module analyzes whether it is a peak traffic period based on real-time traffic data, or based on historical traffic analysis, for example, during long holidays and other time periods, there are many vehicles traveling back and forth. During peak traffic periods, charging stations are only available to pure electric vehicles, and the charging piles are not pushed to hybrid electric vehicles, thus forcing hybrid electric vehicles to meet subsequent mileage by refueling;

[0060] During the peak traffic period, it can also be judged by real-time monitoring of the number of vehicles at the service area entrance or the occupancy rate of charging piles. When the number of vehicles at the entrance exceeds the preset threshold or the occupancy rate of charging piles ≥ 80%, the system automatically enters the peak period mode.

[0061] Specifically, some charging piles in the charging station are single-gun. In order to ensure that all charging piles in the charging station can be scheduled and allocated, when the charging piles are all occupied for single-gun charging piles, the background charging planning module plans the electric vehicle to the parking space on the opposite side of the charging pile with the most charging amount, so as to queue up for vehicle charging.

[0062] Specifically, the background charging planning module is provided with a dynamic power distribution module, a priority adjustment module, and a power limit optimization module;

[0063] The dynamic power distribution module dynamically adjusts the power distribution for each charging pile according to the battery capacity, remaining power, charging demand of the currently charging vehicle, and the overall power load of the charging station:

[0064]

[0065] In the formula: P i The allocated power of the i-th charging pile, G i The battery capacity of the vehicle connected to the i-th charging pile, S i The percentage of the remaining power of the vehicle connected to the i-th charging pile, P total The total power of the charging station, and n is the number of vehicles currently charging in the charging station;

[0066] For vehicles with relatively low remaining power, the priority adjustment module automatically increases their charging priority:

[0067]

[0068] In the formula: Priority i The priority of the i-th charging pile, G max The maximum battery capacity of all vehicles in the current charging station;

[0069] When the total power of the charging station is close to the upper limit, the power limit optimization module automatically reduces the charging power of low-priority vehicles to ensure that high-priority vehicles can obtain sufficient charging power:

[0070]

[0071] In the formula: P′ i Is the power of the i-th charging pile after adjustment, P max The maximum power limit of the charging station, P usedTotal power already used by the current charging station.

[0072] Through the dynamic power distribution algorithm, the system can optimize power according to the real-time needs of vehicles, improve charging efficiency, reduce power waste in the charging station, and meet the charging needs of different vehicles.

[0073] Specifically, the background charging planning module is used to predict the charging demand of each service area in real time and provide input data for the scheduling algorithm:

[0074] Q(t, s) = σ(Conv3D(W conv *[H(t - Δt), T(t), W(s)] + b conv ));

[0075] Where: Q(t, s) is the predicted charging demand of service area s at time t, unit: vehicles per hour, H(t - Δt) is the time series data of the charging amount in the historical Δt period, dimension K×1, where Δt is the size of the time window, T(t) is the real-time traffic flow matrix, dimension N×N, representing the traffic conditions of the road network around the service area, W conv is the three-dimensional convolution kernel weight for spatio-temporal feature extraction, b conv is the bias term of the convolutional layer, and σ is the LeakyReLU activation function, defined as:

[0076]

[0077] W(s) is the weather influence factor, calculated by the cubic polynomial fitting of humidity h and temperature τ:

[0078] W(s) = a0 + a1h + a2τ + a3h 2 + a4τ 2 + a5h 3 + a6τ 3 .

[0079] Through the three-dimensional convolution operation that fuses multi-dimensional spatio-temporal features, the prediction accuracy of the measured data is improved by 23.6% compared with the traditional ARIMA model.

[0080] Specifically, based on the predicted charging demand Q(t, s), the scheduling algorithm optimizes the allocation of charging piles by the dynamic priority scheduling algorithm:

[0081]

[0082] The constraint conditions are:

[0083]

[0084] Where: U is the total number of electric vehicles to be scheduled, V is the total number of charging piles in the charging station, W is the number of supply-demand balance constraints, xuv is a binary decision variable indicating whether vehicle u is assigned to charging pile v, where x uv = 1 means assigned, and x uv = 0 means not assigned;

[0085] The first part of the objective function Minimize the comprehensive cost, including distance cost, electricity cost, and waiting time cost. The second part Punish the imbalance between supply and demand to ensure that the scheduling scheme matches the actual demand;

[0086] The first constraint Ensure that the number of assigned vehicles to each charging pile v does not exceed its maximum capacity;

[0087] The second constraint E u x uv ≥R min Ensure that when vehicle u is assigned to a charging pile, its remaining power meets the minimum charging requirement;

[0088] The third constraint Ensure that the deviation between the scheduling scheme and the predicted demand is within the allowable range.

[0089] For the first time, introduce the supply-demand balance slack variable y w , so that the scheduling scheme remains feasible in case of emergencies, and the measured scheduling success rate is increased by 18.4%.

[0090] c uv is the comprehensive cost of vehicle u assigned to charging pile v, and the calculation formula is:

[0091]

[0092] In the formula: d uv is the distance from vehicle u to charging pile v, d max is the maximum allowable distance in the system, E u is the remaining power of the vehicle, E max is the maximum capacity of the vehicle battery, t wait is the estimated waiting time of vehicle u, t max is the maximum allowable waiting time in the system, α, β, γ are weight coefficients, satisfying α + β + γ = 1, C v is the maximum capacity of charging pile v (i.e., the number of vehicles that can be served simultaneously), R min is the minimum power requirement for vehicle charging, Q(t, s) is the predicted charging demand of service area s at time t provided by the charging demand dynamic prediction model, y wis the slack variable for the supply-demand balance constraint, representing the deviation between the actual allocation and the demand forecast. ∈ is the maximum deviation allowed for the supply-demand balance, and λ is the penalty factor used to control the impact of the supply-demand balance deviation. The calculation formula is as follows:

[0093]

[0094] In the formula: C total is the total capacity of all charging piles in the system.

[0095] Introduce the dynamic penalty factor λ, which is dynamically adjusted according to the real-time demand Q(t, s) and the total system capacity C total to ensure that the supply-demand balance is preferentially satisfied during high-demand periods. α, β, and y can be dynamically adjusted according to real-time traffic data and user preferences. Introduce slack variables to handle the supply-demand imbalance problem and ensure the feasibility of the scheduling scheme in case of emergencies. The cost function c uv adopts a non-linear combination to better meet the multi-objective optimization requirements in the actual scenario.

[0096] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. The patent protection scope of the present invention is subject to the claims. Any equivalent structural changes made by using the content of the specification of the present invention should be equally included in the protection scope of the present invention.

Claims

1. Internet of Things service management system based on regional digital economy, characterized by: It includes a backend charging planning module, multiple charging stations set up in high-speed service areas, and a positioning module and a navigation module set up on electric vehicles; Each charging station is equipped with multiple rows of electric vehicle charging piles, each charging pile is equipped with two charging parking spaces, which are symmetrically arranged on the front and rear sides of the charging pile, and each parking space is equipped with a ground lock. When the electric vehicle is fully charged, the ground lock automatically opens, and when the electric vehicle leaves the charging pile after charging, the ground lock automatically closes. The backend charging planning module is used to plan and allocate the electric vehicle to the vehicle charging pile according to the electric power and location information of the electric vehicle, and pre-store the license plate number of the electric vehicle in the ground lock of the charging pile; When the positioning module detects that the electric vehicle passes through the charging station planned by the background charging planning module, the background charging planning module cancels the license plate number of the electric vehicle pre-stored in the ground lock of the charging pile; The ground lock is provided with a license plate recognition module, which is used to identify the electric vehicle planned to the location by the background charging planning module. When the recognition is correct, the ground lock is automatically opened; If the recognition fails, the navigation module on the electric vehicle will prompt the owner to perform manual verification; When an electric vehicle that is charging at a charging pile suddenly disconnects and stops charging and the electric vehicle drives out of the charging pile, the background charging planning module plans the electric vehicle that was originally navigated to the charging station and is closest to the charging station to enter the charging station, and the planning navigation module navigates the vehicle to the charging pile that suddenly disconnects and stops charging; A redundant communication mechanism is provided between the ground lock and the background charging planning module. When the network is interrupted, the ground lock completes the verification through the locally cached license plate information.

2. The Internet of Things service management system based on regional digital economy according to claim 1 is characterized in that: The background charging planning module analyzes whether it is a peak period of traffic flow based on real-time traffic data. During the peak period of traffic flow, the charging station is only used by pure electric vehicles, and the charging pile is not pushed to hybrid electric vehicles; The peak traffic flow period is determined by real-time monitoring of the number of vehicles entering the service area or the occupancy rate of charging piles. When the number of vehicles entering exceeds a preset threshold or the occupancy rate of charging piles is ≥80%, the system automatically enters the peak mode.

3. The Internet of Things service management system based on regional digital economy according to claim 1 is characterized in that: The charging pile is a single-gun. When the charging piles are all occupied, the background charging planning module plans the electric vehicle to the parking space on the opposite side of the charging pile with the largest charging amount.

4. The Internet of Things service management system based on regional digital economy according to claim 1 is characterized in that: The background charging planning module is provided with a dynamic power allocation module, a priority adjustment module and a power limit optimization module; The dynamic power allocation module dynamically adjusts the power allocation for each charging pile according to the battery capacity, remaining power, charging demand of the current charging vehicle and the overall power load of the charging station: Where: P i The allocated power of the i-th charging pile, G i The battery capacity of the vehicle connected to the i-th charging pile, S i The remaining power percentage of the vehicle connected to the i-th charging pile, P total The total power of the charging station, n is the number of vehicles currently charging in the charging station; For vehicles with low remaining power, the priority adjustment module automatically increases their charging priority: Where: Priority i The priority of the i-th charging pile, G max The maximum battery capacity of all vehicles currently in the charging station; When the total power of the charging station approaches the upper limit, the power limit optimization module automatically reduces the charging power of low-priority vehicles to ensure that high-priority vehicles can obtain sufficient charging power: Where: P′ i is the power of the ith charging pile after adjustment, P max The maximum power limit of the charging station, P used The total power currently used by the charging station.

5. The Internet of Things service management system based on regional digital economy according to claim 1 is characterized in that: The background charging planning module is used to predict the charging demand of each service area in real time and provide input data for the scheduling algorithm: Q(t,s)=σ(Conv3D(W conv *[H(t-Δt),T(t),W(s)]+b conv )); Where: Q(t, s) is the predicted charging demand of service area s at time t, unit: vehicle / hour, H(t-Δt) is the time series data of charging capacity in the historical period Δt, dimension K×1, where Δt is the time window size, T(t) is the real-time traffic flow matrix, dimension N×N, representing the traffic conditions of the road network around the service area, W conv is the 3D convolution kernel weight, used for spatiotemporal feature extraction, b conv is the bias term of the convolutional layer, and σ is the LeakyReLU activation function, which is defined as: W(s) is the weather influencing factor, which is calculated by fitting a cubic polynomial of humidity h and temperature τ: W(s)=a0+a1h+a2τ+a3h 2 +a4τ 2 +a5h 3 +a6τ 3 。 6. The Internet of Things service management system based on regional digital economy according to claim 5 is characterized in that: The scheduling algorithm is based on the predicted charging demand Q(t, s), and the dynamic priority scheduling algorithm optimizes the allocation of charging piles: The constraints are: Where: U is the total number of electric vehicles to be dispatched, V is the total number of charging piles in the charging station, W is the number of supply and demand balance constraints, x uv is a binary decision variable, indicating whether vehicle u is assigned to charging station v, x uv =1 means allocation, x uv =0 means no allocation, c uv The comprehensive cost of assigning vehicle u to charging pile v is calculated as follows: Where: d uv is the distance from vehicle u to charging station v, d max is the maximum allowable distance in the system, E u is the remaining power of the vehicle, E max is the maximum capacity of the vehicle battery, t wait is the expected waiting time of vehicle u, t max is the maximum allowable waiting time in the system, α, β, γ are weight coefficients, satisfying α+β+γ=1, C v is the maximum capacity of the charging pile v, R min The minimum power requirement for charging the vehicle, Q(t, s) is the predicted charging demand of service area s at time t provided by the dynamic prediction model of charging demand, and y w is the slack variable of the supply-demand balance constraint, indicating the deviation between the actual allocation and the demand forecast, ∈ is the maximum deviation allowed by the supply-demand balance, and λ is the penalty factor used to control the impact of the supply-demand balance deviation. The calculation formula is: Where: C total is the total capacity of all charging piles in the system.

Images