City-level intelligent parking integrated management system based on economics principle

Abstract

The invention relates to the technical field of parking integrated management, in particular to a city-level intelligent parking integrated management system based on an economics principle, which comprises a data collection module, a modeling module, a quantum coding module, a quantum application module and a scheme determination module, the modeling module carries out problem modeling, the quantum coding module converts a problem into a QUBO form and creates a price mapping function, the quantum application module solves an optimization problem by using a quantum annealing principle, and the scheme determination module generates and verifies a final price scheme; the innovative process effectively solves the limitation of the traditional system in the aspects of data processing, optimization algorithm, nonlinear relation modeling, dynamic pricing, multi-objective optimization and the like; the system realizes more accurate demand prediction, a more flexible pricing strategy and more efficient resource allocation, and provides a comprehensively optimized solution for urban parking management.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated parking management, and in particular to a city-level intelligent integrated parking management system based on economic principles. Background Art

[0002] Traditional city-level intelligent parking integrated management systems face multiple technical challenges when dealing with complex parking pricing issues.

[0003] First, traditional systems have limitations in optimization algorithms, making it difficult to find the global optimal solution within a reasonable time, especially when faced with large-scale, multi-variable, and multi-constrained complex problems.

[0004] Secondly, traditional systems often use simplified linear models, which are difficult to fully capture the nonlinear relationships and complex economic principles in parking behavior, such as price elasticity and diminishing marginal utility. In terms of pricing strategies, the system usually adopts a relatively static approach, which makes it difficult to achieve flexible dynamic pricing and regional differentiated pricing. At the same time, traditional systems are not effective in dealing with multi-objective optimization problems, and it is difficult to take into account multiple objectives such as resource utilization, user satisfaction and total revenue at the same time.

[0005] Finally, in terms of constraint handling, traditional systems often use simple hard constraint methods, which are difficult to flexibly handle complex constraints, especially when multiple constraints that affect each other need to be considered. These technical problems make it difficult for traditional systems to provide truly optimized solutions for urban parking management. Summary of the invention

[0006] The purpose of the present invention is to provide a city-level intelligent parking integrated management system based on economic principles to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a city-level intelligent parking integrated management system based on economic principles, comprising a data collection module, a modeling module, a quantum coding module, a quantum application module and a solution determination module, wherein: The data collection module collects static basic data and historical statistical data related to urban parking; The modeling module performs problem modeling based on static basic data and historical statistical data, wherein the problem modeling includes the number of regions, the number of time periods, the regional time period price, the constraint condition set and the optimization function; The quantum coding module converts the constraint set and the optimization function into QUBO form to generate a problem matrix; and creates a price mapping function and encodes quantum bits according to the number of regions, the number of time periods, and the regional time period prices; The quantum application module constructs an initial quantum state according to the encoding of the quantum bit; and constructs a problem Hamiltonian according to the problem matrix; uses the Schrödinger equation and the annealing path of the problem Hamiltonian to evolve the initial quantum state into a final quantum state and convert it into a binary result; The scheme determination module decodes the binary result into a measured price scheme according to the price mapping function, and repeatedly executes the quantum application module and the scheme measurement module to generate an optimized price scheme; and verifies and adjusts the optimized price scheme according to the constraint condition set to generate a final price scheme.

[0008] As a further improvement of the present technical solution, the static basic data in the data collection module includes but is not limited to parking facility information and geographic information; historical statistical data includes but is not limited to parking usage records and traffic flow data; dynamic real-time data includes but is not limited to real-time parking space occupancy and real-time road traffic flow, wherein the real-time parking space occupancy is collected in real time through sensors at the parking lot entrances and exits and on the parking spaces, and the collected data includes but is not limited to the occupancy status of each parking space, vehicle entry and exit information, etc.; real-time road traffic flow is collected in real time using video surveillance and induction coil equipment on the road, and the collected data includes but is not limited to vehicle flow, vehicle speed, and congestion level.

[0009] As a further improvement of the technical solution, the modeling module performs problem modeling based on static basic data and historical statistical data, specifically including: City zoning: Based on the geographic information in the static basic data, the city is divided into several parking areas. The geographic locations are grouped by using clustering algorithms and the geographic information data is preprocessed. The clustering algorithm is applied to group the data points according to the preset features. Each data point represents its feature vector in the multidimensional space and is grouped according to the feature similarity to form the regional division result. Time division: Extract parking demand in each period of a day from historical parking data, perform data cleaning and preprocessing to remove abnormal data and fill missing values; use time series analysis technology to smooth the data and extract the main trend of demand changes from the noise; use time series decomposition to decompose the time series into long-term trends, seasonal components and random fluctuations, and further identify the daily rules and peak hours of demand; based on the above analysis results, divide a day into several time periods, and the parking demand in each time period has relatively consistent characteristics; According to the above city division results, the number of regions N is determined, which is a fixed integer; according to the time division results, the number of time periods M is determined; Define a price variable for each region i in each time period j, which is an N×M matrix representing the entire pricing scheme. The regional time period price is expressed as P[i][j], where i represents the index of the region and j represents the index of the time period; The process of establishing a set of constraints in problem modeling is as follows: The constraint set includes price range constraint, time period continuity constraint, regional equilibrium constraint and total revenue target constraint, among which: The price range constraint sets the minimum and maximum price limits for each zone in the form of: Pmin[i]≤P[i][j]≤Pmax[i], where Pmin[i] represents the minimum parking price allowed in the i-th zone, and Pmax[i] represents the maximum parking price allowed in the i-th zone; Time period continuity constraint: limit the price change range between adjacent time periods to avoid drastic price fluctuations. The form is as follows: |P[i][j]-P[i][j+1]|≤ΔPmax, where ΔPmax represents the maximum price change range allowed between adjacent time periods. Regional equilibrium constraint: ensure that the price difference between adjacent regions is within a reasonable range, in the form of |P[i][j]-P[k][j]|≤εmax, where i and k are adjacent regions and εmax represents the maximum price difference allowed between adjacent regions; Total revenue target constraint: Set a minimum total revenue target to ensure the economic feasibility of the pricing plan, in the form of: Σ(i,j)(P[i][j]*D[i][j])≥Rmin, where D[i][j] is the estimated demand and Rmin represents the set minimum total revenue target; The optimization function in problem modeling aims to maximize parking resource utilization and revenue while taking into account user satisfaction. The process of defining the optimization function MaxZ is as follows: MaxZ=w1*(Σ(i,j)U[i][j])-w2*(Σ(i,j)|P[i][j]-P_opt[i][j]|)+w3*R, where U[i][j] is the estimated utilization of area i in time period j, U[i][j] is estimated by the demand function, U[i][j]=f(P[i][j],D[i][j],C[i]), where C[i] is the parking capacity of area i, f is a function fitted based on historical data; P_opt[i][j] is the optimal price calculated based on historical data; R is the total revenue; w1, w2, w3 are weight coefficients used to balance different objectives.

[0010] As a further improvement of the technical solution, the quantum coding module includes a problem matrix generation unit, which converts the constraint set and the optimization function into a QUBO form to generate a problem matrix, where QUBO represents unconstrained quadratic binary optimization; the process of generating the problem matrix is ​​as follows: Convert the optimization function MaxZ into a minimization problem: MinZ=-MaxZ; Convert the constraints into penalty terms and add them to the objective function: Price range constraints: A*Σ(i,j)max(0,Pmin[i]-P[i][j])^2+A*Σ(i,j)max(0,P[i][j]-Pmax[i])^2; Period continuity constraints: B*Σ(i,j)max(0,|P[i][j]-P[i][j+1]|-ΔPmax)^2; Regional equilibrium constraints: C*Σ(i,k,j)max(0,|P[i][j]-P[k][j]|-εmax)^2, where i and k are adjacent regions; Total revenue target constraint: D*max(0,Rmin-Σ(i,j)(P[i][j]*D[i][j]))^2; The above A, B, C, and D are penalty coefficients used to adjust the weights of each constraint; Add the penalty term to MinZ to obtain the objective function of the QUBO problem. The objective function QUBO_Z is as follows: QUBO_Z=MinZ+penalty term, expand QUBO_Z and organize it into a quadratic form: QUBO_Z=x^TQx+c, where x is the decision variable vector containing all P[i][j]; Q is the problem matrix containing the coefficients of all quadratic terms; and c is the constant term.

[0011] As a further improvement of the technical solution, the quantum coding module includes a coding unit, and the coding unit encodes the quantum bits according to the number of regions, the number of time periods, and the regional time period price, specifically including: For each time period j of each region i, encode the price P[i][j] into binary form. Assuming that b bits are used to represent each price, then P[i][j]=Σ(k=0tob-1)2^k*q[i][j][k], where q[i][j][k] is the qubit representing the kth binary bit of P[i][j], to represents the range from the starting value to the ending value, and k represents the index of the binary bit; the total number of qubits required is N*M*b, where N is the number of regions and M is the number of time periods; Create a price mapping function to map each q[i][j][k] to the variables in the QUBO problem. The price mapping function is as follows: x[index]=q[i][j][k], where index=i*M*b+j*b+k; this mapping function converts the three-dimensional q[i][j][k] into the one-dimensional x[index] for representation in the QUBO matrix; According to this encoding scheme, all P[i][j] in QUBO_Z are re-expressed and replaced with the corresponding binary representation, resulting in a QUBO problem based entirely on quantum bits, as follows: QUBO_Z=x^TQx+c, where x is now a vector of length NMb, representing all the quantum bit encodings; Q is a (NMb)x(NMb) matrix, representing the coefficients of all quadratic terms; and c is a constant term.

[0012] As a further improvement of the technical solution, the quantum application module constructs an initial quantum state according to the encoding of the quantum bit; and constructs the problem Hamiltonian according to the problem matrix; uses the Schrödinger equation and the annealing path of the problem Hamiltonian to evolve the initial quantum state into the final quantum state and convert it into a binary result, specifically including: Constructing the initial quantum state: Based on the quantum bit code generated by the encoding unit, an initial quantum state is created. The initial state contains NMb quantum bits, where N is the number of regions, M is the number of time periods, and b is the number of bits representing each price. The initial state is set to a uniform superposition state, indicating that each quantum bit is in a superposition state of 0 and 1 with equal probability. Construct the problem Hamiltonian: Using the QUBO problem matrix Q and the constant term c generated by the problem matrix generation unit, the module constructs the problem Hamiltonian H_p, which contains all the information of the problem; Design annealing path: define a time-varying Hamiltonian H(t) to represent the evolution from the initial Hamiltonian H_i to the problem Hamiltonian H_p. The initial Hamiltonian H_i is usually chosen as the transverse magnetic field Hamiltonian, and H(t) is a linear combination of H_i and H_p, whose coefficients vary with time. Quantum state evolution: Using the Schrödinger equation, the module simulates the evolution of the quantum state from the initial state to the final state, which starts at time t=0 and lasts until the total annealing time T. In this process, the quantum system gradually changes from the initial uniform superposition state to a state closer to the optimal solution of the problem. Measuring the final quantum state: At the end of the annealing process, that is, t=T, the final quantum state is measured. This measurement will collapse the quantum state into a classical state, that is, obtain the specific 0 or 1 value of each quantum bit q[i][j][k], which is the final quantum state.

[0013] As a further improvement of the technical solution, the solution determination module decodes the binary result into a measurement price solution according to the price mapping function, and repeatedly executes the quantum application module and the solution measurement module to generate an optimized price solution; and verifies and adjusts the optimized price solution according to the constraint condition set to generate a final price solution, which specifically includes: Using the price mapping function defined by the encoding unit, decode the binary result back to the original price variable P[i][j], and reassemble the b binary bits of each region i and period j into the corresponding price value; The module repeatedly executes the quantum application module and the measurement process multiple times, and each execution generates a potential price scheme. By comparing the objective function values ​​of these schemes, i.e., the values ​​of MaxZ, the best scheme is selected as the candidate optimized price scheme. Verify the candidate optimization price plan and check whether it meets all the preset constraint sets according to the constraint set in the modeling module, as follows: Price range constraint: ensure that each P[i][j] is between Pmin[i] and Pmax[i]; Period continuity constraint: check whether the price change in adjacent periods does not exceed ΔPmax; Regional equilibrium constraint: verify whether the price difference between adjacent regions is within the range of εmax; Total revenue target constraint: calculate the total revenue and ensure it is not less than Rmin; If the candidate optimized price solution does not meet certain constraints, adjustments are made, including truncating or scaling prices that are out of range, smoothing price changes between adjacent time periods, adjusting price differences between adjacent areas, and increasing or decreasing prices in certain areas or time periods; After verification and adjustment, the module generates a final price plan that meets all constraints. The plan is an N×M matrix, where each element P[i][j] represents the optimal parking price in area i during time period j.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This urban-level intelligent parking integrated management system based on economic principles achieves supply and demand balance through a dynamic pricing mechanism P[i][j], and uses the estimated utilization rate U[i][j] and demand D[i][j] to reflect the price elasticity theory, which fully reflects the core principles of economics; the constraints Pmin[i], Pmax[i] and ΔPmax take into account the diminishing marginal utility, and the optimization function MaxZ includes externality considerations, making the pricing strategy more comprehensive; the use of differentiated pricing for region i and time period j reflects the discriminatory pricing strategy and effectively improves resource utilization efficiency; the use of weight coefficients w1, w2, and w3 reflects the Pareto optimality idea, and the εmax constraint reflects the game relationship between adjacent regions and promotes overall equilibrium; the total revenue target Rmin takes into account opportunity costs and economies of scale to ensure the economic feasibility of the system; the entire process uses static and historical decision-making, which reflects the principles of information economics, and at the same time incorporates behavioral economics ideas through considerations of user satisfaction, making the pricing strategy more humane and acceptable.

[0015] 2. The price mapping function in this city-level intelligent parking comprehensive management system based on economic principles plays an important role in multiple key links and significantly improves the efficiency and accuracy of the system. First, in the quantum coding stage, the price mapping function converts the continuous price variable P[i][j] into a discrete quantum bit representation q[i][j][k], realizing an effective mapping from classical problems to quantum problems. This mapping not only retains the structure of the original problem, but also enables the problem to take advantage of quantum computing. Secondly, after the quantum computing is completed, the price mapping function is used to convert the quantum state measurement results back to the actual price plan, ensuring the actual availability of the calculation results. In addition, by repeatedly executing the quantum application module and using the price mapping function, the system can generate multiple candidate plans and select the optimal solution from them, greatly improving the quality and stability of the final plan.

[0016] 3. The multiple applications of constraint sets in this urban-level intelligent parking integrated management system based on economic principles greatly enhance the practicality and reliability of the system; in the initial problem modeling stage, the constraint set defines the boundary of the problem to ensure that the generated solution meets the actual operational needs; when the problem is converted into QUBO form, the constraints are cleverly converted into penalty terms, so that the quantum algorithm can automatically consider these constraints in the solution process; finally, in the solution determination stage, the constraint set is again used to verify and adjust the optimized price plan, further ensuring the feasibility of the final plan; this multi-level constraint application not only ensures the actual executability of the optimization results, but also improves the system's adaptability to complex actual situations, so that the generated parking price plan not only meets the requirements of mathematical optimization, but also meets various constraints of actual operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall module of the present invention; Figure 2 Schematic diagram of the quantum coding module unit of the present invention.

[0018] In the figure: 100, data collection module; 200, modeling module; 300, quantum coding module; 301, problem matrix generation unit; 302, coding unit; 400, quantum application module; 500, solution determination module. DETAILED DESCRIPTION

[0019] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0021] The following are some definitions of terms: Static basic data: refers to relatively stable basic data such as parking facility information (such as parking lot location, capacity, and type) and geographic information (such as urban area division, road distribution, and distribution of commercial and residential areas); Historical statistics: including parking usage records in the past period (such as parking time, duration, and fees) and historical traffic flow data on major roads; Constraint set: including price range constraints, time period continuity constraints, regional equilibrium constraints, and total revenue target constraints, etc., which are used to limit the reasonable range of parking pricing schemes; Optimization function: a mathematical expression for maximizing parking resource utilization and revenue while taking into account user satisfaction, in the form of MaxZ; Price mapping function: a function that converts the three-dimensional quantum bit representation q[i][j][k] into the one-dimensional QUBO problem variable x[index]; Encoding of qubits: the process of converting the price P[i][j] of each region and period into binary form and representing it with qubits; Initial quantum state: a uniform superposition state created based on quantum bit encoding, where each quantum bit is in a superposition state of 0 and 1 with equal probability; Final quantum state: After the quantum annealing process, the quantum state that is closer to the optimal solution to the problem can be measured to obtain a specific binary result.

[0022] Next, see Figure 1-Figure 2 The present invention provides a technical solution: a city-level intelligent parking integrated management system based on economic principles, including a data collection module 100, a modeling module 200, a quantum coding module 300, a quantum application module 400 and a solution determination module 500.

[0023] The data collection module 100 collects static basic data and historical statistical data related to urban parking, as follows: Static basic data includes, but is not limited to, parking facility information (e.g., location, capacity, type, and other basic information of parking lots and roadside parking spaces), geographic information (e.g., urban area division, main road distribution, commercial areas, residential areas, and other geographic information); Historical statistical data include, but are not limited to, parking usage records (for example, each parking lot management system obtains parking records for a certain period in the past, including parking time, duration, fees, etc.), traffic flow data (for example, historical traffic flow data on major roads).

[0024] The modeling module 200 performs problem modeling based on static basic data and historical statistical data, wherein the problem modeling includes the number of regions, the number of time periods, the price of the region and time period, a set of constraint conditions, and an optimization function. The construction of the number of regions, the number of time periods, and the price of the region and time period in the modeling process specifically includes: Urban zoning: Based on the geographic information in the static basic data (such as commercial density, residential density, traffic flow and the distribution of existing parking facilities, etc.), the city is divided into several parking areas. The geographical locations are grouped by using clustering algorithms. First, the geographic information data is preprocessed, such as removing noise and outliers; then, the clustering algorithm (such as K-means) is applied to group the data points according to the preset features. Each data point represents its feature vector in the multidimensional space, and the data points are grouped according to the feature similarity to form the regional division result; such a zoning method ensures that each parking area has relatively consistent characteristics in terms of commercial activities, residents' needs and traffic characteristics, and provides a reasonable geographical unit for subsequent parking pricing; Time division: Extract parking demand in each time period of a day from historical parking data, perform data cleaning and preprocessing to remove abnormal data and fill missing values; then, use time series analysis techniques, such as moving average method, to smooth the data and extract the main trend of demand changes from the noise; next, use time series decomposition (such as season-trend decomposition or Fourier transform) to decompose the time series into long-term trends, seasonal components and random fluctuations, and further identify the daily patterns and peak hours of demand; based on these analysis results, divide a day into several time periods, such as morning peak, working hours, lunch time, evening peak, night, etc., and the parking demand in each time period has relatively consistent characteristics; the final time division results can reflect the demand characteristics of different time periods, providing a scientific basis for the detailed analysis of static basic data in the time dimension; The number of regions N is determined according to the above city division results, which is a fixed integer; the number of time periods M is determined according to the time division results, for example, a day may be divided into 6-8 time periods; Define a price variable for each region i in each time period j, which is an N×M matrix representing the entire pricing scheme. The regional time period price is expressed as (P[i][j]), where i represents the index of the region and j represents the index of the time period.

[0025] The process of establishing a set of constraints in problem modeling is as follows: The constraint set includes price range constraint, time period continuity constraint, regional equilibrium constraint and total revenue target constraint, among which: The price range constraint sets the minimum and maximum price limits for each zone in the form of Pmin[i]≤P[i][j]≤Pmax[i], where Pmin[i] represents the minimum parking price allowed in the i-th zone and Pmax[i] represents the maximum parking price allowed in the i-th zone; Time period continuity constraint: limit the price change range between adjacent time periods to avoid drastic price fluctuations. The format is: |P[i][j]-P[i][j+1]|≤ΔPmax, where ΔPmax represents the maximum price change range allowed between adjacent time periods. Regional equilibrium constraint: ensure that the price difference between adjacent regions is within a reasonable range, in the form of |P[i][j]-P[k][j]|≤εmax, where i and k are adjacent regions, and εmax represents the maximum price difference allowed between adjacent regions; Total revenue target constraint: Set a minimum total revenue target to ensure the economic feasibility of the pricing plan, in the form of: Σ(i,j)(P[i][j]*D[i][j])≥Rmin, where D[i][j] is the estimated demand and Rmin represents the set minimum total revenue target.

[0026] The optimization function in problem modeling aims to maximize parking resource utilization and revenue while taking into account user satisfaction. The process of defining the optimization function MaxZ is as follows: MaxZ=w1*(Σ(i,j)U[i][j])-w2*(Σ(i,j)|P[i][j]-P_opt[i][j]|)+w3*R, where U[i][j] is the estimated utilization of area i in time period j, U[i][j] is estimated by the demand function, U[i][j]=f(P[i][j],D[i][j],C[i]), where C[i] is the parking capacity of area i, f is a function fitted based on historical data; P_opt[i][j] is the optimal price calculated based on historical data; R is the total revenue; w1, w2, w3 are weight coefficients used to balance different objectives.

[0027] Through this problem modeling process, the complex urban parking pricing problem is transformed into a mathematical model with clear parameters, constraints and optimization objectives.

[0028] The problem matrix generation unit 301 in the quantum coding module 300 converts the constraint set and the optimization function into QUBO form to generate a problem matrix, where QUBO represents unconstrained quadratic binary optimization; the process of generating the problem matrix is ​​as follows: Convert the optimization function MaxZ into a minimization problem: MinZ=-MaxZ; Convert the constraints into penalty terms and add them to the objective function: Price range constraints: A*Σ(i,j)max(0,Pmin[i]-P[i][j])^2+A*Σ(i,j)max(0,P[i][j]-Pmax[i])^2; Period continuity constraints: B*Σ(i,j)max(0,|P[i][j]-P[i][j+1]|-ΔPmax)^2; Regional equilibrium constraints: C*Σ(i,k,j)max(0,|P[i][j]-P[k][j]|-εmax)^2, where i and k are adjacent regions; Total revenue target constraint: D*max(0,Rmin-Σ(i,j)(P[i][j]*D[i][j]))^2; The above A, B, C, and D are penalty coefficients used to adjust the weights of each constraint; Adding these penalty terms to MinZ, we get the objective function of the QUBO problem. The objective function QUBO_Z is as follows: QUBO_Z=MinZ+penalty term, expand QUBO_Z and organize it into a quadratic form: QUBO_Z=x^TQx+c, where x is the decision variable vector containing all P[i][j]; Q is the problem matrix containing the coefficients of all quadratic terms; and c is the constant term.

[0029] The encoding unit 302 in the quantum encoding module 300 encodes the quantum bits according to the number of regions, the number of time periods, and the regional time period price, specifically including: For each time period j of each region i, encode the price P[i][j] into binary form. Assuming that b bits are used to represent each price, then P[i][j]=Σ(k=0tob-1)2^k*q[i][j][k], where q[i][j][k] is the qubit representing the kth binary bit of P[i][j], to represents the range from the starting value to the ending value, and k represents the index of the binary bit; the total number of qubits required is N*M*b, where N is the number of regions and M is the number of time periods; Create a price mapping function that maps each q[i][j][k] to the variable price mapping function in the QUBO problem as follows: x[index]=q[i][j][k], where index=i*M*b+j*b+k; this mapping function converts the three-dimensional q[i][j][k] into the one-dimensional x[index], which is convenient for representation in the QUBO matrix; According to this encoding scheme, all P[i][j] in QUBO_Z are re-expressed and replaced with the corresponding binary representation, resulting in a QUBO problem based entirely on quantum bits, as follows: QUBO_Z=x^TQx+c, where x is now a vector of length NMb, representing all the quantum bit encodings; Q is a (NMb)x(NMb) matrix, representing the coefficients of all quadratic terms; and c is a constant term.

[0030] Through the process in the quantum coding module 300, the original parking pricing problem is converted into a QUBO form suitable for solving by the quantum annealing algorithm, and the encoding of quantum bits is completed. The result can be directly input into the quantum annealing hardware or simulator for optimization and solution.

[0031] The quantum application module 400 constructs an initial quantum state according to the quantum bit encoding in the encoding unit 302; and constructs the problem Hamiltonian according to the problem matrix; and uses the Schrödinger equation and the annealing path of the problem Hamiltonian to evolve the initial quantum state into a final quantum state and convert it into a binary result, specifically including: Constructing an initial quantum state: Based on the quantum bit code generated by the encoding unit 302, the module creates an initial quantum state, which contains NMb quantum bits, where N is the number of regions, M is the number of time periods, and b is the number of bits representing each price; the initial state is set to a uniform superposition state, which means that each quantum bit is in a superposition state of 0 and 1 with equal probability; Constructing the problem Hamiltonian: using the QUBO problem matrix Q and the constant term c generated by the problem matrix generation unit 301, the module constructs the problem Hamiltonian H_p, which contains all the information of the problem, including the optimization goal and the constraints; Design annealing path: define a time-varying Hamiltonian H(t) to represent the evolution from the initial Hamiltonian H_i to the problem Hamiltonian H_p. The initial Hamiltonian H_i is usually chosen as the transverse magnetic field Hamiltonian, and H(t) is a linear combination of H_i and H_p, whose coefficients vary with time. Quantum state evolution: Using the Schrödinger equation, the module simulates the evolution of the quantum state from the initial state to the final state, which starts at time t=0 and lasts until the total annealing time T. In this process, the quantum system gradually changes from the initial uniform superposition state to a state closer to the optimal solution of the problem. Measuring the final quantum state: At the end of the annealing process (t=T), the module measures the final quantum state. This measurement collapses the quantum state into a classical state, that is, obtaining the specific 0 or 1 value of each quantum bit q[i][j][k], that is, the final quantum state.

[0032] The solution determination module 500 decodes the binary result into a measured price solution according to the price mapping function, and repeatedly executes the quantum application module 400 and the solution measurement module to generate an optimized price solution; and verifies and adjusts the optimized price solution according to the constraint condition set to generate a final price solution, which specifically includes: Decoding the binary result back to the original price variable P[i][j] using the price mapping function defined by the encoding unit 302, which involves recombining the b binary bits for each region i and time period j into the corresponding price value; In order to improve the reliability and quality of the results, the module repeatedly executes the quantum application module 400 and the measurement process multiple times, and each execution generates a potential price scheme; by comparing the objective function values ​​of these schemes (i.e., the values ​​of MaxZ), the best scheme is selected as the candidate optimized price scheme; The candidate optimization price solution is verified, and according to the constraint condition set in the modeling module 200, it is checked whether all preset constraint conditions are met, as follows: Price range constraint: ensure that each P[i][j] is between Pmin[i] and Pmax[i]; Period continuity constraint: check whether the price change in adjacent periods does not exceed ΔPmax; Regional equilibrium constraint: verify whether the price difference between adjacent regions is within the range of εmax; Total revenue target constraint: calculate the total revenue and ensure it is not less than Rmin; If the candidate optimized price solution does not meet certain constraints, adjustments are made, including truncating or scaling prices that are out of range, smoothing price changes between adjacent time periods, adjusting price differences between adjacent areas, and increasing or decreasing prices in certain areas or time periods; After verification and adjustment, the module generates a final price plan that meets all constraints. This plan is an N×M matrix, where each element P[i][j] represents the optimal parking price for area i in time period j. The module calculates the objective function value MaxZ of the final price scheme and evaluates its impact on parking resource utilization, user satisfaction and total revenue. The evaluation result is used to compare the schemes obtained with different running times.

[0033] The process of this system embodies several core economic principles. It achieves supply and demand balance through the dynamic pricing mechanism P[i][j], and uses the estimated utilization rate U[i][j] and demand D[i][j] to reflect the price elasticity theory. The constraints Pmin[i], Pmax[i] and ΔPmax take into account the diminishing marginal utility, while the optimization function MaxZ includes externality considerations; the use of differentiated pricing for region i and time period j reflects the discriminatory pricing strategy. The use of weight coefficients w1, w2, and w3 reflects the Pareto optimality idea, and the εmax constraint reflects the game relationship between adjacent regions; the total revenue target Rmin takes into account opportunity costs and economies of scale. The entire process uses static, historical, and dynamic data to make decisions, reflecting the principles of information economics, and at the same time incorporates behavioral economics through the consideration of user satisfaction.

[0034] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A city-level intelligent parking integrated management system based on economic principles, characterized in that: The method comprises a data collection module (100), a modeling module (200), a quantum coding module (300), a quantum application module (400) and a solution determination module (500), wherein: The data collection module (100) collects static basic data and historical statistical data related to urban parking; The modeling module (200) performs problem modeling based on static basic data and historical statistical data, wherein the problem modeling includes the number of regions, the number of time periods, the price of the region and time period, a set of constraint conditions and an optimization function; The quantum coding module (300) converts the constraint condition set and the optimization function into a QUBO form to generate a problem matrix; and creates a price mapping function and encodes quantum bits according to the number of regions, the number of time periods, and the region-time period prices; The quantum application module (400) constructs an initial quantum state according to the encoding of the quantum bit; and constructs a problem Hamiltonian according to the problem matrix; and uses the Schrödinger equation and the annealing path of the problem Hamiltonian to evolve the initial quantum state into a final quantum state, and converts it into a binary result; The solution determination module (500) decodes the binary result into a measured price solution according to the price mapping function, and repeatedly executes the quantum application module (400) and the solution measurement module to generate an optimized price solution; and verifies and adjusts the optimized price solution according to the constraint condition set to generate a final price solution.

2. The city-level intelligent parking integrated management system based on economic principles according to claim 1 is characterized in that: The static basic data in the data collection module (100) includes, but is not limited to, parking facility information and geographic information; the historical statistical data includes, but is not limited to, parking usage records and traffic flow data; the dynamic real-time data includes, but is not limited to, real-time parking space occupancy and real-time road traffic flow, wherein the real-time parking space occupancy is collected in real time through sensors at the parking lot entrance and exit and on the parking spaces, and the collected data includes, but is not limited to, the occupancy status of each parking space, vehicle entry and exit information, etc.; Real-time road traffic flow is collected in real time using video surveillance and induction coil equipment on the road. The collected data includes but is not limited to vehicle flow, vehicle speed, and congestion level.

3. The city-level intelligent parking integrated management system based on economic principles according to claim 1 is characterized in that: The modeling module (200) performs problem modeling based on static basic data and historical statistical data, specifically including: City zoning: Based on the geographic information in the static basic data, the city is divided into several parking areas. The geographic locations are grouped by using clustering algorithms and the geographic information data is preprocessed. The clustering algorithm is applied to group the data points according to the preset features. Each data point represents its feature vector in the multidimensional space and is grouped according to the feature similarity to form the regional division result. Time division: Extract parking demand in each period of a day from historical parking data, perform data cleaning and preprocessing to remove abnormal data and fill missing values; use time series analysis technology to smooth the data and extract the main trend of demand changes from the noise; use time series decomposition to decompose the time series into long-term trends, seasonal components and random fluctuations, and further identify the daily rules and peak hours of demand; based on the above analysis results, divide a day into several time periods, and the parking demand in each time period has relatively consistent characteristics; According to the above city division results, the number of regions N is determined, which is a fixed integer; according to the time division results, the number of time periods M is determined; Define a price variable for each region i in each time period j, which is an N×M matrix representing the entire pricing scheme. The regional time period price is expressed as P[i][j], where i represents the index of the region and j represents the index of the time period; The process of establishing a set of constraints in problem modeling is as follows: The constraint set includes price range constraint, time period continuity constraint, regional equilibrium constraint and total revenue target constraint, among which: The price range constraint sets the minimum and maximum price limits for each zone in the form of: Pmin[i]≤P[i][j]≤Pmax[i], where Pmin[i] represents the minimum parking price allowed in the i-th zone, and Pmax[i] represents the maximum parking price allowed in the i-th zone; Time period continuity constraint: limit the price change range between adjacent time periods to avoid drastic price fluctuations. The form is as follows: |P[i][j]-P[i][j+1]|≤ΔPmax, where ΔPmax represents the maximum price change range allowed between adjacent time periods. Regional equilibrium constraint: ensure that the price difference between adjacent regions is within a reasonable range, in the form of |P[i][j]-P[k][j]|≤εmax, where i and k are adjacent regions and εmax represents the maximum price difference allowed between adjacent regions; Total revenue target constraint: Set a minimum total revenue target to ensure the economic feasibility of the pricing plan, in the form of: Σ(i,j)(P[i][j]*D[i][j])≥Rmin, where D[i][j] is the estimated demand and Rmin represents the set minimum total revenue target; The optimization function in problem modeling aims to maximize parking resource utilization and revenue while taking into account user satisfaction. The process of defining the optimization function MaxZ is as follows: MaxZ=w1*(Σ(i,j)U[i][j])-w2*(Σ(i,j)|P[i][j]-P_opt[i][j]|)+w3*R, where U[i][j] is the estimated utilization of area i in time period j, U[i][j] is estimated by the demand function, U[i][j]=f(P[i][j],D[i][j],C[i]), where C[i] is the parking capacity of area i, f is a function fitted based on historical data; P_opt[i][j] is the optimal price calculated based on historical data; R is the total revenue; w1, w2, w3 are weight coefficients used to balance different objectives.

4. The city-level intelligent parking integrated management system based on economic principles according to claim 3 is characterized in that: The quantum coding module (300) comprises a problem matrix generating unit (301), wherein the problem matrix generating unit (301) converts the constraint condition set and the optimization function into a QUBO form to generate a problem matrix, wherein QUBO represents unconstrained quadratic binary optimization; the process of generating the problem matrix is ​​as follows: Convert the optimization function MaxZ into a minimization problem: MinZ=-MaxZ; Convert the constraints into penalty terms and add them to the objective function: Price range constraints: A*Σ(i,j)max(0,Pmin[i]-P[i][j])^2+A*Σ(i,j)max(0,P[i][j]-Pmax[i])^2; Period continuity constraints: B*Σ(i,j)max(0,|P[i][j]-P[i][j+1]|-ΔPmax)^2; Regional equilibrium constraints: C*Σ(i,k,j)max(0,|P[i][j]-P[k][j]|-εmax)^2, where i and k are adjacent regions; Total revenue target constraint: D*max(0,Rmin-Σ(i,j)(P[i][j]*D[i][j]))^2; The above A, B, C, and D are penalty coefficients used to adjust the weights of each constraint; Add the penalty term to MinZ to obtain the objective function of the QUBO problem. The objective function QUBO_Z is as follows: QUBO_Z=MinZ+penalty term, expand QUBO_Z and organize it into a quadratic form: QUBO_Z=x^TQx+c, where x is the decision variable vector containing all P[i][j]; Q is the problem matrix containing the coefficients of all quadratic terms; and c is the constant term.

5. The city-level intelligent parking integrated management system based on economic principles according to claim 4 is characterized in that: The quantum coding module (300) comprises a coding unit (302), wherein the coding unit (302) performs quantum bit coding according to the number of regions, the number of time periods, and the regional time period price, specifically comprising: For each time period j of each region i, encode the price P[i][j] into binary form. Assuming that b bits are used to represent each price, then P[i][j]=Σ(k=0tob-1)2^k*q[i][j][k], where q[i][j][k] is the qubit representing the kth binary bit of P[i][j], to represents the range from the starting value to the ending value, and k represents the index of the binary bit; the total number of qubits required is N*M*b, where N is the number of regions and M is the number of time periods; Create a price mapping function to map each q[i][j][k] to the variables in the QUBO problem. The price mapping function is as follows: x[index]=q[i][j][k], where index=i*M*b+j*b+k; this mapping function converts the three-dimensional q[i][j][k] into the one-dimensional x[index] for representation in the QUBO matrix; According to this encoding scheme, all P[i][j] in QUBO_Z are re-expressed and replaced with the corresponding binary representation, resulting in a QUBO problem based entirely on quantum bits, as follows: QUBO_Z=x^TQx+c, where x is now a vector of length NMb, representing all the quantum bit encodings; Q is a (NMb)x(NMb) matrix, representing the coefficients of all quadratic terms; and c is a constant term.

6. The city-level intelligent parking integrated management system based on economic principles according to claim 5 is characterized in that: The quantum application module (400) constructs an initial quantum state according to the encoding of the quantum bits; And based on the problem matrix, construct the problem Hamiltonian; Using the Schrödinger equation and the annealing path of the problem Hamiltonian, the initial quantum state is evolved into the final quantum state and converted into a binary result, including: Constructing an initial quantum state: based on the quantum bit code generated by the encoding unit (302), creating an initial quantum state, the initial state comprising NMb quantum bits, where N is the number of regions, M is the number of time periods, and b is the number of bits representing each price; the initial state is set to a uniform superposition state, indicating that each quantum bit is in a superposition state of 0 and 1 with equal probability; Constructing the problem Hamiltonian: using the QUBO problem matrix Q and the constant term c generated by the problem matrix generation unit (301), the module constructs the problem Hamiltonian H_p, which contains all the information of the problem; Design annealing path: define a time-varying Hamiltonian H(t) to represent the evolution from the initial Hamiltonian H_i to the problem Hamiltonian H_p. The initial Hamiltonian H_i is usually chosen as the transverse magnetic field Hamiltonian, and H(t) is a linear combination of H_i and H_p, whose coefficients vary with time. Quantum state evolution: Using the Schrödinger equation, the module simulates the evolution of the quantum state from the initial state to the final state, which starts at time t=0 and lasts until the total annealing time T. In this process, the quantum system gradually changes from the initial uniform superposition state to a state closer to the optimal solution of the problem. Measuring the final quantum state: At the end of the annealing process, that is, t=T, the final quantum state is measured. This measurement will collapse the quantum state into a classical state, that is, obtain the specific 0 or 1 value of each quantum bit q[i][j][k], which is the final quantum state.

7. The city-level intelligent parking integrated management system based on economic principles according to claim 5 is characterized in that: The solution determination module (500) decodes the binary result into a measurement price solution according to the price mapping function, and repeatedly executes the quantum application module (400) and the solution measurement module to generate an optimized price solution; And according to the constraint set, the optimized price plan is verified and adjusted to generate the final price plan, including: Using the price mapping function defined by the encoding unit (302), the binary result is decoded back to the original price variable P[i][j], and the b binary bits of each region i and time period j are reassembled into the corresponding price value; The module repeatedly executes the quantum application module (400) and the measurement process multiple times, and each execution generates a potential price plan; by comparing the objective function values ​​of these plans, that is, the values ​​of MaxZ, the best plan is selected as a candidate optimized price plan; The candidate optimization price solution is verified by checking whether all preset constraint sets are satisfied according to the constraint set in the modeling module (200), as follows: Price range constraint: ensure that each P[i][j] is between Pmin[i] and Pmax[i]; Period continuity constraint: check whether the price change in adjacent periods does not exceed ΔPmax; Regional equilibrium constraint: verify whether the price difference between adjacent regions is within the range of εmax; Total revenue target constraint: calculate the total revenue and ensure it is not less than Rmin; If the candidate optimized price solution does not meet certain constraints, adjustments are made, including truncating or scaling prices that are out of range, smoothing price changes between adjacent time periods, adjusting price differences between adjacent areas, and increasing or decreasing prices in certain areas or time periods; After verification and adjustment, the module generates a final price plan that meets all constraints. The plan is an N×M matrix, where each element P[i][j] represents the optimal parking price in area i during time period j.

Images