Method for Quantifying Uncertainty of Charging Station Load Based on Spatiotemporal Traffic Flow Prediction

Abstract

The present invention discloses a method for quantifying the uncertainty of charging station load based on spatio-temporal traffic flow prediction. The method includes collecting traffic flow data from sensor nodes on both sides of the road, including traffic flow data and average vehicle speed data; constructing a traffic flow uncertainty prediction model based on a quantile spatio-temporal network, and using the model and traffic flow data to predict traffic flow uncertainty; considering factors such as charging station electricity price, traffic flow, and road congestion, constructing an improved charging station queuing model, and converting the uncertainty prediction result of traffic flow into the uncertainty result of charging station load. The present invention fully excavates the spatio-temporal coupling characteristics of traffic flow data, effectively improves the accuracy of traffic flow uncertainty prediction. On the other hand, the present invention comprehensively considers the influence of factors such as traffic flow and charging electricity price on the charging station load, and further improves the accuracy of charging station load uncertainty prediction.

Description

Technical Field

[0001] The present invention relates to a method for quantifying the uncertainty of charging station load based on spatio-temporal traffic flow prediction, belonging to the field of power systems. Background Art

[0002] The rapid development of electric vehicles (EVs) is changing the global traffic pattern, with the core driving force being the urgent pursuit of carbon emission reduction and clean alternative energy. This trend has led to a sharp increase in the number of electric vehicles on the road and their charging demands, thus posing a severe challenge to the stability and power supply capacity of the power grid. Moreover, the charging load of electric vehicles has a high degree of non-stationarity, and traditional deterministic prediction is difficult to accurately describe its load characteristics. Uncertainty prediction can quantify and predict the uncertainty of electric vehicles, which plays a crucial supporting role in the dispatching and control decisions of the power grid.

[0003] Existing research on electric vehicle load prediction does not fully consider the influence of traffic flow, road congestion coefficient, electricity price, etc. on the queuing process. On the other hand, existing research on traffic flow uncertainty prediction cannot fully consider the spatio-temporal correlation relationship between the traffic flows of each charging station node, which restricts the accuracy of charging load uncertainty prediction. Therefore, in order to effectively improve the accuracy of charging station load uncertainty, the present invention proposes a method for quantifying the uncertainty of charging station load based on spatio-temporal traffic flow prediction. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method for quantifying the uncertainty of charging station load based on spatio-temporal traffic flow prediction. The present invention can fully consider the spatio-temporal correlation characteristics of traffic flow data for traffic flow uncertainty prediction. On this basis, it can consider the charging electricity price, traffic flow, and road congestion factors of the charging station to achieve an accurate conversion of traffic flow uncertainty into charging station load uncertainty. To achieve the above purpose, the present invention is implemented by the following technical solutions, specifically including the following steps:

[0005] (1) Collect traffic flow data from sensor nodes on both sides of the road, including traffic flow data and average vehicle speed data;

[0006] (2) Construct and train a traffic flow uncertainty prediction model based on a quantile spatio-temporal network, and use the model and traffic flow data to predict traffic flow uncertainty;

[0007] (3) Construct a charging station electric vehicle queuing model considering the charging electricity price, traffic flow, and road congestion factors of the charging station, and convert the traffic flow uncertainty prediction result into a charging station load uncertainty prediction result.

[0008] Further, the step (1) includes:

[0009] Collect traffic flow data from sensor nodes on both sides of the road, including traffic flow volume data and average vehicle speed data, specifically expressed as

[0010] X = [x1, x2, …, x T ∈ R T×N×H , where X is the total traffic flow data, and x1, x2, …, x T represent the traffic flow data at t = 1, 2, …, T, N is the number of nodes, T is the time length, H is the number of traffic flow data characteristics of each node, and R represents the set of real numbers.

[0011] Furthermore, step (2) includes:

[0012] (2.1) Construct a traffic flow uncertainty prediction model based on the quantile spatio-temporal network, including constructing a traffic flow data time series feature extraction module, a traffic flow data spatial feature extraction module, a traffic flow data spatio-temporal fusion module, and a quantile-based traffic flow uncertainty prediction quantification module:

[0013] (2.1.1) Construct a traffic flow data time series feature extraction module: This module is composed of an LSTM network. At time t, the time series feature extraction module obtains h t , and the h t at different times are concatenated to obtain the time series feature X T ,

[0014] (2.1.2) Construct a traffic flow data spatial feature extraction module: This module consists of a static feature extraction module and a dynamic feature extraction module.

[0015] The static feature extraction module is specifically expressed as:

[0016]

[0017] where Y S represents the static spatial feature, x′ represents the traffic flow data, I N represents the unit diagonal matrix, W (l) and W (l+1) represent the weight matrices of the l-th and l + 1-th neural networks, σ is the non-linear activation function, LayerNorm(·) represents layer normalization, and A is the adjacency matrix, which is used to represent the road distance and connection relationship between sensor nodes.

[0018] The dynamic feature extraction module is specifically expressed as:

[0019] Y D = LayerNorm(σ(Sx'W (l)) + x′′)

[0020] Among them, Y D represents the dynamic spatial feature, S is the dynamic spatial correlation matrix, S = softmax(V s ·σ((x'W1)W2(W3x′) T +b s ))), where x′ represents traffic flow data, V s , W1, W2, W3, b s are learnable weights,

[0021] Fuse the static feature extraction module and the dynamic feature extraction module to obtain the output Y of the traffic flow data spatial feature extraction module, which is specifically expressed as Y = FC(FC(Y s ) + FC(Y D ))), where FC() represents the fully connected neural network,

[0022] (2.1.3) Construct the traffic flow data spatio-temporal fusion module:

[0023] The traffic flow data spatio-temporal fusion module is specifically expressed as:

[0024]

[0025] x″′ = W O concat(c1, c2,..., c n )

[0026] Among them, W qi , W ki , W vi is the learnable weight parameter of the i-th head, Q i , K i , V i represent data matrices, c i represents the output of the i-th head, d m , d q , d k , d v are the corresponding matrix dimensions, softmax is the activation function, W o represents the output weight, concat represents the concatenation function, x″′ represents the output of the traffic flow data spatio-temporal fusion module, () T represents the matrix transpose, is the set of real numbers,

[0027] (2.1.4) Traffic flow uncertainty prediction quantization module:

[0028] To quantify the uncertainty of traffic flow, the spatio-temporal fusion module of traffic flow data is transformed into quantile prediction results at different quantile levels, which are specifically expressed as:

[0029]

[0030] In the formula, is the quantile prediction result of the i-th sample at the r-th quantile level κ r under, m i，j refers to the j-th output result of the i-th sample of the spatio-temporal fusion module of traffic flow data.

[0031] Therefore, the goal of traffic flow uncertainty prediction quantization is expressed as minimizing the loss, which is expressed as:

[0032]

[0033]

[0034] Among them, y c，i represents the true value of the i-th sample, κ r is the quantile level, η is an arbitrarily small positive number, is the index aggregation of the data set, is the index set of the quantiles, |·| represents the cardinality function, is the importance degree of the quantiles, L (r) represents the loss of the traffic flow uncertainty prediction quantization model at the quantile level κ r under.

[0035] (2.2) Train the traffic flow uncertainty prediction model based on the quantile spatio-temporal network in (2.1), and use the trained model and data to predict the traffic flow uncertainty.

[0036] (2.2.1) Train the traffic flow uncertainty prediction model based on the quantile spatio-temporal network in (2.1):

[0037] Use traffic flow data to train the uncertainty prediction model, and the training goal is to minimize the loss described in (2.1.4).

[0038] (2.2.2) Use the trained model and data to predict the traffic flow uncertainty and achieve the prediction of the traffic flow uncertainty.

[0039] Furthermore, the step (3) includes:

[0040] (3.1) Construct a charging station electric vehicle queuing model considering the charging electricity price, traffic flow and road congestion factors of the charging station:

[0041] Let \(N(t)\) be the number of electric vehicles in the charging station at time \(t\). The electric vehicle queuing model conforms to a birth-death process with the state space \(H = \{1, 2, 3, \cdots, E\}\), where the birth rate \(\lambda\) k and the death rate \(\mu\) k are respectively expressed as:

[0042]

[0043] where \(C\) is the number of charging piles in the charging station, \(E\) is the maximum capacity of the charging station, \(k\) is the number of electric vehicles in the current charging station, \(\lambda\) represents the arrival rate of electric vehicles, \(\mu\) refers to the departure rate, which is the average number of electric vehicles leaving the charging station within a fixed time, \(\alpha=(a + b - d)\), \(\beta=(\delta + b - d)\), \(\alpha\) is the sensitivity coefficient of electric vehicle users who want to charge with respect to the queue length, \(\delta\) is the sensitivity coefficient of electric vehicle users who want to leave the charging station with respect to the queue length, \(b\) represents the electricity price, and \(d\) represents the degree of road congestion.

[0044] And let the probability that the number of electric vehicles in the charging station is \(k\) be \(k = 0, 1, \cdots, E\), and its corresponding steady-state equation is expressed as:

[0045] (\(\lambda\) k +\(\mu\) k )\(p\) k =\(\lambda\) k-1 \(p\) k-1 +\(\lambda\) k+1 \(p\) k+1

[0046]

[0047] (3.2) Using the model constructed in (3.1) to convert the traffic flow uncertainty result into the charging station load uncertainty result:

[0048] Solving the model described in (3.1), we get:

[0049]

[0050]

[0051] where, \(G\) represents the average number of electric vehicles in the charging station when the arrival rate is \(\lambda\).

[0052] After calculating the average number of electric vehicles in the charging station for each time period, the average charging load for each time period is calculated and expressed as:

[0053] \(P\) cha_all =\(G * P\) ch

[0054] where \(P\)ch is the rated charging power of the charging pile, P cha_all is the load power of the charging station, that is, the uncertainty of the charging load of the charging station. Since the arrival rate λ is calculated, that is, the result of traffic flow uncertainty, what is calculated by this formula is also the uncertainty of the charging load of the charging station.

[0055] An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for quantifying the uncertainty of the charging station load based on spatio-temporal traffic flow prediction as described above is implemented.

[0056] A computer-readable storage medium stores computer instructions thereon. When the computer instructions are executed by a processor, the method for quantifying the uncertainty of the charging station load based on spatio-temporal traffic flow prediction as described above is implemented.

[0057] Compared with the prior art, the beneficial effects achieved by the method for quantifying the uncertainty of the charging station load based on spatio-temporal traffic flow prediction provided by the embodiments of the present invention include:

[0058] (1) The traffic flow uncertainty prediction algorithm proposed by the present invention respectively constructs a time feature extraction module, a space feature extraction module, and a spatio-temporal fusion module, which can fully explore the spatio-temporal correlation characteristics of the traffic flow data of the charging station nodes and improve the accuracy of power generation power aggregation.

[0059] (2) Different from the prior art, the present invention constructs a charging station queuing model considering the influence of factors such as the charging electricity price, traffic flow, and road congestion of the charging station, and can realize the accurate conversion of the traffic flow uncertainty prediction result to the charging station load uncertainty prediction result considering various real factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is a flowchart of the method for quantifying the uncertainty of the charging station load based on spatio-temporal traffic flow prediction provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the protection scope of the present invention.

[0062] Embodiment: As Figure 1 shown, this embodiment provides a quantification of the uncertainty of the charging station load based on quantile spatio-temporal traffic flow prediction, including:

[0063] (1) Collect traffic flow data from sensor nodes on both sides of the road, including traffic flow data and average vehicle speed data;

[0064] (2) Construct and train a traffic flow uncertainty prediction model based on the quantile spatio-temporal network, and use this model and traffic flow data to predict traffic flow uncertainty;

[0065] (3) Construct a charging station electric vehicle queuing model considering charging electricity prices, traffic flow, and road congestion factors at the charging station, and convert the traffic flow uncertainty prediction results into charging station load uncertainty prediction results.

[0066] The specific steps include:

[0067] Step 1: Collect traffic flow data from sensor nodes on both sides of the road, including traffic flow data and average vehicle speed data, specifically expressed as

[0068] X = [x1, x2,..., x T ∈ R T×N×H , where x is the total traffic flow data, and x1, x2,..., x T represent traffic flow data at t = 1, 2,..., T, N is the number of nodes, T is the time length, H is the number of traffic flow data characteristics of each node, and R represents the set of real numbers.

[0069] Step 2: Construct and train a traffic flow uncertainty prediction model based on the quantile spatio-temporal network, and use this model and traffic flow data to predict traffic flow uncertainty;

[0070] (2.1) Construct a traffic flow uncertainty prediction model based on the quantile spatio-temporal network, including constructing a traffic flow data time series feature extraction module, a traffic flow data spatial feature extraction module, a traffic flow data spatio-temporal fusion module, and a quantile-based traffic flow uncertainty prediction quantization module:

[0071] (2.1.1) Construct a traffic flow data time series feature extraction module: This module consists of an LSTM network. At time t, the time series feature extraction module obtains h t , and splices the h t at different times to obtain the time series feature X T ,

[0072] (2.1.2) Construct a traffic flow data spatial feature extraction module: This module consists of a static feature extraction module and a dynamic feature extraction module.

[0073] The static feature extraction module is specifically expressed as:

[0074]

[0075] Among them, Y S represents the static spatial feature, x' represents the traffic flow data, I Ndenotes the unit diagonal matrix, W (l) and W (l+1) denote the weight matrices of the l-th and (l + 1)-th layers of the neural network, σ is the non-linear activation function, LayerNorm(·) represents layer normalization, and A is the adjacency matrix, which is used to represent the road distances and connection relationships between sensor nodes.

[0076] The dynamic feature extraction module is specifically expressed as:

[0077] Y D = LayerNorm(σ(Sx'W (l) ) + x′)

[0078] where Y D represents the dynamic spatial features, S is the dynamic spatial correlation matrix, S = softmax(V s ·σ((x'W1)W2(W3x′) T + b s ))), x′ represents the traffic flow data, and V s , W1, W2, W3, b s are learnable weights.

[0079] Fuse the static feature extraction module and the dynamic feature extraction module to obtain the output Y of the traffic flow data spatial feature extraction module, which is specifically expressed as Y = FC(FC(Y s ) + FC(Y D )), where FC() represents the fully connected neural network.

[0080] (2.1.3) Construct the traffic flow data spatio-temporal fusion module:

[0081] The traffic flow data spatio-temporal fusion module is specifically expressed as:

[0082]

[0083] x″′ = W O concat(c1, c2,..., c n )

[0084] where W qi , W ki , W vi are the learnable weight parameters of the i-th head, Q i , K i , V i denote the data matrices, c i denotes the output of the i-th head, and d m , d q , d k , d vis the corresponding matrix dimension, softmax is the activation function, and W o represents the output weight, concat represents the concatenation function, x″′ represents the output of the traffic flow data spatio-temporal fusion module, and () T represents matrix transpose, is the set of real numbers,

[0085] (2.1.4) Traffic flow uncertainty prediction quantization module:

[0086] is to quantify the uncertainty of traffic flow and convert the traffic flow data spatio-temporal fusion module into quantile prediction results at different quantile levels, specifically expressed as:

[0087]

[0088] In the formula, is the quantile prediction result of the i-th sample at the r-th quantile level κ r under, and m i，j refers to the j-th output result of the i-th sample of the traffic flow data spatio-temporal fusion module.

[0089] Therefore, the goal of traffic flow uncertainty prediction quantization is expressed as minimizing the loss, expressed as:

[0090]

[0091] Among them, y c，i represents the true value of the i-th sample, κ r is the quantile level, η is an arbitrarily small positive number, is the index aggregation of the dataset, is the index set of quantiles, |·| represents the cardinality function, is the importance degree of the quantile, and L (r) represents the loss of the traffic flow uncertainty prediction quantization model at the quantile level κ r under.

[0092] (2.2) Train the traffic flow uncertainty prediction model based on the quantile spatio-temporal network in (2.1), and use the trained model and data to predict the traffic flow uncertainty,

[0093] (2.2.1) Train the traffic flow uncertainty prediction model based on the quantile spatio-temporal network in (2.1):

[0094] Use traffic flow data to train the uncertainty prediction model, and the training goal is to minimize the loss described in (2.1.4),

[0095] (2.2.2) Use the trained model and data to predict the uncertainty of traffic flow and achieve the prediction of traffic flow uncertainty.

[0096] Step 3: Construct a charging station electric vehicle queuing model considering the charging electricity price, traffic flow, and road congestion factors of the charging station, and convert the traffic flow uncertainty prediction result into the charging station load uncertainty prediction result.

[0097] (3.1) Construct a charging station electric vehicle queuing model considering the charging electricity price, traffic flow, and road congestion factors of the charging station:

[0098] Let N(t) be the number of electric vehicles in the charging station at time t. The electric vehicle queuing model conforms to a birth-death process with the state space H = {1, 2, 3,...E}, where the birth rate λ k and the death rate μ k are respectively expressed as:

[0099]

[0100] Among them, C is the number of charging piles in the charging station, E is the maximum capacity of the charging station, k is the number of electric vehicles in the current charging station, λ represents the arrival rate of electric vehicles, μ refers to the departure rate, which is the average number of electric vehicles leaving the charging station within a fixed time, α = (a + b - d), β = (δ + b - d), α is the sensitivity coefficient of electric vehicle users who want to charge regarding the queue length, δ is the sensitivity coefficient of electric vehicle users who want to leave the charging station regarding the queue length, b represents the electricity price, and d represents the degree of road congestion.

[0101] And let the probability that the number of electric vehicles in the charging station is k be k = 0, 1,...E, and its corresponding steady-state equation is expressed as:

[0102] (λ k + μ k )p k = λ k-1 p k-1 + λ k+1 p k+1

[0103]

[0104] (3.2) Use the model constructed in (3.1) to convert the traffic flow uncertainty result into the charging station load uncertainty result:

[0105] Solve the model described in (3.1) to obtain:

[0106]

[0107] Among them, Let \(G\) denote the average number of electric vehicles in the charging station when the arrival rate is \(\lambda\).

[0108] After calculating the average number of electric vehicles in the charging station for each time period, the average charging load for each time period is calculated as follows:

[0109] \(P\) cha_all \(= G\times P\) ch

[0110] where \(P\) ch is the rated charging power of the charging pile, and \(P\) cha_all is the load power of the charging station, that is, the uncertainty of the charging load of the charging station. Since the arrival rate \(\lambda\) is obtained by calculation, that is, the result of traffic flow uncertainty, what is calculated by this formula is also the uncertainty of the charging load of the charging station.

[0111] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and deformations can be made, and these improvements and deformations should also be regarded as the protection scope of the present invention.

Claims

1. A charging station load uncertainty quantification method based on spatiotemporal traffic flow prediction, characterized in that: The following steps are involved: (1) Collect traffic flow data from sensor nodes on both sides of the road, including traffic volume data and average vehicle speed data; (2) Construct and train a traffic flow uncertainty prediction model based on quantile spatiotemporal networks, and use this model and traffic flow data to predict traffic flow uncertainty; (3) Construct an electric vehicle queuing model for charging stations that takes into account charging electricity prices, traffic flow, and road congestion, and convert the traffic flow uncertainty prediction results into charging station load uncertainty prediction results; Step (1) is specifically as follows: Traffic flow data is collected from sensor nodes on both sides of the road, including traffic flow data and average vehicle speed data, which is specifically expressed as X = [x1, x2, ..., x T ]∈R T×N×H , where X is the total traffic flow data, x1, x2, …, x T represents the traffic flow data at t=1, 2, ..., T, N is the number of nodes, T is the time length, H is the number of traffic flow data features of each node, and R represents a real number set; Step (2) is specifically as follows: (2.1) Constructing a traffic flow uncertainty prediction model based on quantile spatiotemporal network, including constructing a traffic flow data temporal feature extraction module, a traffic flow data spatial feature extraction module, a traffic flow data spatiotemporal fusion module and a traffic flow uncertainty prediction quantification module based on quantiles: (2.1.1) Constructing a traffic flow data temporal feature extraction module: This module is composed of an LSTM network. At time t, the temporal feature extraction module obtains h t , h at different times t Splice to get the timing characteristics X T , (2.1.2) Constructing the spatial feature extraction module of traffic flow data: This module consists of a static feature extraction module and a dynamic feature extraction module. The static feature extraction module is specifically expressed as: Among them, Y S represents static spatial features, x′ represents traffic flow data, I N represents the unit diagonal matrix, W (l) and W (l+1) represents the weight matrix of the lth and l+1th layers of the neural network, σ is the nonlinear activation function, LayerNorm(·) represents layer normalization, A is the adjacency matrix, which is used to represent the road distance and connection relationship between sensor nodes, The dynamic feature extraction module is specifically expressed as: Y D =LayerNorm(σ(Sx'W(l))+x′) where Y D represents the dynamic spatial features, S is the dynamic spatial association matrix, S = softmax(V s ·σ((x′W1)W2(W3x′)T+b s ), x′ represents traffic flow data, V s ,W1,W2,W3,b s are learnable weights, The static feature extraction module and the dynamic feature extraction module are integrated to obtain the output Y of the traffic flow data spatial feature extraction module, which is specifically expressed as: Y = FC (FC (Y s )+FC(Y D )), where FC() represents a fully connected neural network, (2.1.3) Constructing the spatiotemporal fusion module of traffic flow data: The traffic flow data spatiotemporal fusion module is specifically expressed as: x″′=W O concat(c1,c2,...,c n ) Among them, W qi , W ki , W vi It is the learnable weight parameter of the i-th head, Q i , K i , V i represents the data matrix, c i represents the output of the i-th head, d m , d q , d k , d v is the corresponding matrix dimension, softmax is the activation function, W o represents the output weight, concat represents the concatenation function, x″′ represents the output of the spatiotemporal fusion module of traffic flow data, () T represents the matrix transpose, is the set of real numbers, (2.1.4) Traffic flow uncertainty prediction quantification module: In order to quantify the uncertainty of traffic flow, the spatiotemporal fusion module of traffic flow data is converted into quantile prediction results at different quantile levels, which can be specifically expressed as follows: In the formula, is the i-th sample at the r-th quantile level κ r Quantile prediction results under m i，j Refers to the jth output result of the i-th sample of the traffic flow data spatiotemporal fusion module, To this end, the goal of traffic flow uncertainty prediction quantification is to minimize the loss, which is expressed as: Among them, y c，i represents the true value of the i-th sample, κ r is the quantile level, η is an arbitrarily small positive number, The index aggregation for the dataset, is the index set of quantiles, |·| represents the cardinality function, is the importance of the quantile, L (r) Indicates that at the quantile level κ r Quantifying model losses for traffic flow prediction under uncertainty, (2.2) Train the traffic flow uncertainty prediction model based on quantile spatiotemporal network in (2.1), and use the trained model and data to predict traffic flow uncertainty. (2.2.1) Train the traffic flow uncertainty prediction model based on quantile spatiotemporal network in (2.1): The uncertainty prediction model is trained using traffic flow data, and the training objective is to minimize the loss described in (2.1.4). (2.2.2) Use the trained model and data to predict traffic flow uncertainty and realize the prediction of traffic flow uncertainty.

2. The charging station load uncertainty quantification method based on spatiotemporal traffic flow prediction according to claim 1 is characterized in that: Step (3) is specifically as follows: (3.1) Construct a charging station electric vehicle queuing model that takes into account charging station electricity prices, traffic flow, and road congestion: Let N(t) be the number of electric vehicles in the charging station at time t. The electric vehicle queuing model conforms to the birth and death process with the state space H = {1, 2, 3, ... E}, where the birth rate λ k and extinction rate μ k Respectively expressed as: Where C is the number of charging piles in the charging station, E is the maximum capacity of the charging station, k is the number of electric vehicles in the current charging station, λ is the arrival rate of electric vehicles, μ is the departure rate, which refers to the average number of electric vehicles leaving the charging station within a fixed time, α = (a + bd), β = (δ + bd), α is the sensitivity coefficient of electric vehicle users who want to charge to the length of the queue, δ is the sensitivity coefficient of electric vehicle users who want to leave the charging station to the length of the queue, b is the electricity price, d is the degree of road congestion, And let the probability that the number of electric vehicles in the charging station is k be k=0,1,...E, and its corresponding steady-state equation is expressed as: (3.2) Using the model constructed in (3.1), the traffic flow uncertainty results are converted into charging station load uncertainty results: Solving the model described in (3.1), we obtain: in, G represents the average number of electric vehicles in the charging station when the arrival rate is λ, After calculating the average number of electric vehicles at the charging station in each time period, the average charging load in each time period is calculated, which is expressed as: P cha_all =G*P ch Among them, P ch is the rated charging power of the charging pile, P cha_all is the load power of the charging station.

3. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for quantifying uncertainty of charging station load based on spatiotemporal traffic flow prediction as described in any one of claims 1 to 2 above is implemented.

4. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, a charging station load uncertainty quantification method based on spatiotemporal traffic flow prediction as described in any one of claims 1-2 is implemented.

Images