Capacity configuration method for road heterogeneous energy complementary power generation system

Abstract

The present application relates to the field of road engineering and new energy technology, and specifically provides a capacity configuration method of a road heterogeneous energy complementary power generation system, comprising: obtaining traffic flow data and meteorological data of a target road section; constructing a heterogeneous energy flow fluctuation model reflecting the combined fluctuation characteristics of piezoelectric output and photovoltaic output, and defining output power variance and flexibility deficiency as fluctuation indicators; taking the minimization of annualized comprehensive cost and the minimization of output power fluctuation as double objectives, taking piezoelectric unit layout density, photovoltaic installed capacity and energy storage configuration capacity as decision variables, and under the constraint conditions including road land and investment budget, a multi-objective optimization algorithm is used to solve the Pareto optimal solution set; the joint probability scenarios of traffic flow and illumination are generated through Monte Carlo simulation, and the reliability of the optimal solution set is evaluated to obtain the capacity configuration parameters. The present application realizes the design change from the maximization of total power generation to the minimization of output fluctuation.

Description

Technical Field

[0001] This invention belongs to the field of road engineering and new energy technology, and specifically relates to a capacity configuration method for a road heterogeneous energy complementary power generation system. Background Technology

[0002] In road infrastructure, piezoelectric power generation utilizes vehicle loads to generate electricity, characterized by high instantaneous power and fast response speed; photovoltaic power generation utilizes solar energy, with relatively stable output but significantly affected by day / night cycles and weather. Both naturally coexist in road scenarios, forming a heterogeneous energy complementary power generation system.

[0003] However, existing capacity allocation methods suffer from the following technical problems: First, there is a lack of quantitative allocation methods tailored to the heterogeneous power output characteristics of piezoelectric and photovoltaic systems. Piezoelectric systems exhibit pulse characteristics, while photovoltaic systems exhibit stable fluctuation characteristics, with time scales differing by several orders of magnitude. Existing methods often rely on conventional microgrid configuration experience and fail to establish quantitative relationships of energy flow specific to roads. Second, road spatial resource constraints (such as the available area of ​​road surface and roadside land) are not incorporated into the configuration model, making it difficult to implement the configuration results. Third, there is a lack of reliability verification indicators oriented towards road load characteristics. The reliability requirements of loads such as lighting, monitoring, charging piles, and emergency response vary significantly, and existing single indicators cannot reflect these differentiated needs. Therefore, there is an urgent need to construct a capacity allocation method that can reflect the unique energy flow characteristics of roads, incorporate spatial resource constraints, and address the differentiated needs of loads. Summary of the Invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present invention provides a capacity configuration method for a road heterogeneous energy complementary power generation system to solve the above-mentioned technical problems.

[0005] This invention provides a capacity configuration method for a road-based heterogeneous energy complementary power generation system, comprising: Obtain traffic flow data and weather data for the target road segment; Based on the traffic flow data and meteorological data, a heterogeneous energy flow fluctuation model reflecting the joint fluctuation characteristics of piezoelectric power output and photovoltaic power output is constructed. Based on the output of the model, the output power variance and flexibility deficit are determined as indicators for evaluating power generation fluctuation. With the dual objectives of minimizing annualized comprehensive cost and minimizing the aforementioned indicators, and with piezoelectric unit deployment density, photovoltaic installed capacity, and energy storage configuration capacity as decision variables, a multi-objective optimization algorithm is used to solve for the candidate capacity configuration parameters under constraints including road land use constraints and investment budget constraints. A joint probability scenario of traffic flow and illumination is generated through Monte Carlo simulation. The reliability of each candidate capacity configuration parameter is evaluated through the joint probability scenario, and the capacity configuration parameter that meets the preset reliability threshold is output. The capacity configuration parameters include the piezoelectric unit deployment density, the photovoltaic installed capacity, and the energy storage configuration capacity.

[0006] In one optional implementation, traffic flow data and meteorological data for the target road segment are acquired, including: Obtain the axle load spectrum distribution, vehicle speed distribution, lane coverage, and vehicle arrival rate from the traffic flow data; Obtain the probability distribution of light intensity, effective sunshine duration, and dynamic shading coefficient from the meteorological data.

[0007] In an optional implementation, acquiring traffic flow data and meteorological data for the target road segment further includes: The axle load spectrum distribution, the vehicle speed distribution, and the vehicle arrival rate are obtained by a dynamic weighing system or traffic flow detectors deployed on the target road section. The lane coverage rate is calculated and determined based on the total number of lanes in the target road segment and the planned number of lanes for the piezoelectric unit deployment; The light intensity probability distribution and the effective sunshine duration are obtained by collecting meteorological station or satellite remote sensing data near the target road section; The dynamic shading coefficient is obtained by measuring the road direction, the height of surrounding features, and the solar azimuth angle on site, combined with pre-calculation using solar radiation analysis software or by establishing a shading coefficient database using a three-dimensional model.

[0008] In an optional implementation, based on the traffic flow data and meteorological data, a heterogeneous energy flow fluctuation model reflecting the joint fluctuation characteristics of piezoelectric and photovoltaic power output is constructed, including: Based on the randomness of traffic flow data, a probability distribution function for piezoelectric output power is established; Based on the probability of sunshine in meteorological data, a probability distribution function for photovoltaic output power is established; Based on the probability distribution function of the piezoelectric output power and the probability distribution function of the photovoltaic output power, a system total power fluctuation model of the combined piezoelectric and photovoltaic output power is constructed.

[0009] In an optional implementation, based on the randomness of traffic flow data, a probability distribution function for the piezoelectric output power is established, including: Axle load spectrum distribution, vehicle speed distribution, lane coverage, and vehicle arrival rate are extracted from the traffic flow data. Based on the empirical relationship between single vehicle piezoelectric energy and axle load and vehicle speed calibrated by indoor loading tests, and combined with the lane coverage rate, a single vehicle piezoelectric energy model is established. Based on the vehicle arrival rate, the probability distribution of the number of vehicles arriving within a time period is determined. Combined with the single vehicle piezoelectric energy model, the expectation and variance of the piezoelectric output power within the time period are calculated, thereby obtaining the probability distribution function of the piezoelectric output power.

[0010] In an optional implementation, a probability distribution function for photovoltaic output power is established based on the sunshine probability in meteorological data, including: Historical light intensity data are extracted from the meteorological data, and a probability distribution model of the original light intensity is obtained by fitting. The probability distribution model uses a Beta distribution to describe the distribution pattern of light intensity. A dynamic shading coefficient is extracted from the meteorological data. The dynamic shading coefficient is calculated based on the road direction, the height of surrounding features, and the solar azimuth angle, and is used to characterize the proportion of light intensity attenuated due to shading. The probability distribution model of the original illumination intensity is corrected using the dynamic shading coefficient to obtain the probability distribution of the effective illumination intensity after shading attenuation. Based on the output characteristics of photovoltaic modules, the probability distribution of the effective light intensity is converted into a probability distribution function of photovoltaic output power.

[0011] In an optional implementation, a system total power fluctuation model for the combined piezoelectric and photovoltaic outputs is constructed based on the probability distribution function of the piezoelectric output power and the probability distribution function of the photovoltaic output power, including: The probability distribution function of the piezoelectric output power and the probability distribution function of the photovoltaic output power within the same time period are convolved to obtain the probability distribution function of the total power. Based on the probability distribution function of the total power, the time-series fluctuation characteristics of the total power are calculated, and the time-series fluctuation characteristics include the output power variance; Based on the power balance relationship between total power and load demand, the flexibility deficit is calculated, which is used to characterize the degree of risk of not being able to meet load demand or absorb excess power.

[0012] In an optional implementation, the constraints include: Road land constraints include: the piezoelectric unit deployment area is limited by the excavable area of ​​the target road section and the lane coverage rate, wherein the lane coverage rate is the ratio of the number of lanes where piezoelectric units are actually deployed to the total number of lanes; the photovoltaic module deployment area is limited by the available area of ​​the roadside slope, sound barrier, service area building roof or parking lot sunshade of the target road section. Investment budget constraints include: the sum of the purchase and installation costs of piezoelectric units, photovoltaic modules, and energy storage systems shall not exceed the preset investment budget ceiling.

[0013] In an optional implementation, a multi-objective optimization algorithm is used to solve for the candidate capacity configuration parameters, including: A non-dominated sorting genetic algorithm with an elitist strategy is adopted as the multi-objective optimization algorithm. The decision variables are encoded to generate an initial population, including the piezoelectric unit deployment density, photovoltaic installed capacity, and energy storage configuration capacity. Based on the annual time-series simulation of the target road segment, the annualized comprehensive cost and output power fluctuation index of the configuration scheme corresponding to each individual in the population are calculated. Individuals in the population are sorted and selected according to non-dominated sorting and crowding distance, and offspring populations are generated through crossover and mutation operations; The population iteration is performed iteratively until the preset termination condition is met, and the final Pareto front solution set is output as a candidate capacity configuration parameter.

[0014] In an optional implementation, a joint probabilistic scenario of traffic flow and illumination is generated through Monte Carlo simulation. The reliability of each candidate capacity configuration parameter is evaluated using this joint probabilistic scenario, and the capacity configuration parameters that meet a preset reliability threshold are output, including: Based on historical meteorological data and historical traffic flow data, various typical operating conditions and the probability of occurrence of each condition are determined. The typical operating conditions include at least normal weather conditions, continuous rainy conditions, and holiday congestion conditions. Based on the probability of occurrence of each working condition, the type of working condition for each simulation is determined by random sampling using the Monte Carlo method; When sampling normal weather conditions or continuous rainy conditions, a first-order Markov chain model is used to generate the corresponding lighting scene sequence; when sampling holiday congestion conditions, a non-homogeneous Poisson process is used and the traffic flow scene sequence is generated by scaling up according to the historical peak ratio. Align the lighting scene sequence and the traffic flow scene sequence generated under the same working condition according to the time step to form a joint probability scene; Repeat the above sampling and generation steps to obtain multiple joint probability scenarios; For each candidate capacity configuration parameter, the system is simulated under various joint probability scenarios. The proportion of time steps in which the system satisfies supply and demand balance and voltage fluctuation does not exceed a preset threshold is counted out of the total number of simulation steps. This proportion is used as the flexibility supply guarantee rate of the configuration scheme. The configuration scheme that achieves a flexible supply guarantee rate that meets the preset reliability threshold will be used as the final output capacity configuration parameter.

[0015] The beneficial effects of this invention are as follows: The capacity configuration method for road-based heterogeneous energy complementary power generation systems provided by this invention establishes a heterogeneous energy flow fluctuation model reflecting the joint fluctuation characteristics of piezoelectric pulse output and stable photovoltaic output. Using output power variance and flexibility deficit as fluctuation indicators, it achieves a design shift from maximizing total power generation to minimizing output fluctuation, thus mitigating power fluctuations of heterogeneous energy sources at the source. By incorporating constraints such as road land use, investment budget, autonomy rate, curtailment rate, and emergency backup into the capacity configuration model, the optimization results truly meet the spatial limitations and economic requirements of road engineering. The introduction of a flexibility supply guarantee rate evaluation index based on Monte Carlo simulation enables the quantification of power supply reliability for extreme conditions such as continuous rain and holiday congestion, meeting the differentiated reliability requirements of different loads such as lighting, monitoring, and emergency response. Through multi-objective optimization to solve the Pareto front, it provides designers with a quantitative decision-making basis between cost and fluctuation, effectively improving the scientific nature and engineering feasibility of road-based heterogeneous energy complementary power generation system planning.

[0016] Furthermore, the design principle of this invention is reliable, the structure is simple, and it has a very wide range of application prospects. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0019] Figure 2 This is a flowchart of a capacity configuration and evaluation method according to an embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram illustrating the construction principle of a heterogeneous energy flow fluctuation model according to an embodiment of the present invention.

[0021] Figure 4 This is a structural block diagram of a capacity configuration device according to an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0024] The capacity configuration method for the road heterogeneous energy complementary power generation system provided in this embodiment of the invention is executed by a computer device, and correspondingly, the capacity configuration system for the road heterogeneous energy complementary power generation system runs in the computer device.

[0025] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention. The order of the steps in this flowchart may be changed, and some steps may be omitted, depending on different requirements.

[0026] like Figure 1 As shown, the method includes: S1. Obtain traffic flow data and meteorological data for the target road segment; S2. Based on the traffic flow data and meteorological data, a heterogeneous energy flow fluctuation model reflecting the joint fluctuation characteristics of piezoelectric output and photovoltaic output is constructed, and the output power variance and flexibility deficit are determined based on the model output as indicators for evaluating power generation fluctuation. S3. With the dual objectives of minimizing annualized comprehensive cost and minimizing the aforementioned indicators, and with piezoelectric unit deployment density, photovoltaic installed capacity, and energy storage configuration capacity as decision variables, a multi-objective optimization algorithm is used to solve for the candidate capacity configuration parameters under constraints including road land use constraints and investment budget constraints. S4. Generate a joint probability scenario of traffic flow and illumination through Monte Carlo simulation, evaluate the reliability of each candidate capacity configuration parameter through the joint probability scenario, and output the capacity configuration parameter that meets the preset reliability threshold. The capacity configuration parameters include the piezoelectric unit deployment density, the photovoltaic installed capacity, and the energy storage configuration capacity.

[0027] Please refer to Figure 2 The capacity configuration and evaluation method for road-domain heterogeneous energy complementary power generation systems provided by this invention includes the following contents.

[0028] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0029] S101. Collect road scene parameters and extract features for the target road segment.

[0030] The system acquires axle load spectrum distribution, vehicle speed distribution, lane coverage, and vehicle arrival rate from traffic flow data. The axle load spectrum distribution, vehicle speed distribution, and vehicle arrival rate are obtained through dynamic weighing systems or traffic flow detectors deployed on the target road segment. The dynamic weighing system can use piezoelectric quartz sensors or bending plate sensors, embedded in the road structure layer. It measures axle load, vehicle speed, axle type, and vehicle type classification in real time as vehicles pass. The axle load spectrum distribution and vehicle speed distribution are obtained through long-term statistical analysis, while the vehicle arrival rate is calculated at 15-minute intervals. If traffic flow detectors (such as loop detectors, radar detectors, or video analysis equipment) are already installed on the road segment, they can also be used as supplementary or verification data. Lane coverage is calculated based on the total number of lanes on the target road segment and the planned number of lanes for piezoelectric units. For example, if a road segment has six lanes in both directions and piezoelectric units are planned to be deployed in two heavy-load lanes, the lane coverage rate is 1 / 3.

[0031] S102. Obtain the probability distribution of light intensity, effective sunshine duration, and dynamic shading coefficient from meteorological data.

[0032] The probability distribution of light intensity and effective sunshine duration were obtained by collecting historical hourly or minute-by-minute data from meteorological stations near the target road section (preferably national-level meteorological stations within 20km of the road section). If nearby stations are unavailable, satellite remote sensing data (such as NASA or European Centre for Medium-Range Weather Forecasts data) can be used, supplemented by on-site portable meteorological stations for calibration. Based on at least 10 years of historical sunshine data, a Beta distribution was used to fit the probability distribution of light intensity, with distribution parameters varying seasonally. Effective sunshine duration was determined based on meteorological station sunshine meter records or by a sustained light intensity exceeding 120 W / m². 2 The dynamic shading coefficient is obtained through statistical analysis of the time period. It is calculated by measuring the road direction, the height of surrounding features (such as buildings, mountains, and sound barriers), and the solar azimuth angle on-site, combined with pre-calculation using solar radiation analysis software (such as Ecotect and PVsyst), or by creating a 3D model through UAV oblique photography, thus generating a shading coefficient database. This database stores the dynamic shading coefficient at different times according to season and hourly granularity, allowing the photovoltaic power output model to call upon and correct it in real time.

[0033] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and its specific implementation will be described in a non-limiting manner. Please refer to [the relevant documentation]. Figure 3 .

[0034] S201: Establish the probability distribution function of piezoelectric output power Based on the randomness of traffic flow data, a probability distribution function for piezoelectric output power is established. Specifically, the vehicle arrival process is described using a non-homogeneous Poisson process, dividing a 24-hour day into 96 time periods (each 15 minutes). The vehicle arrival rate λ(t) within each time period is fitted using historical traffic flow data, and the daily variation of the arrival rate can be fitted using a Fourier series. The single-vehicle piezoelectric energy model is calibrated based on indoor MTS loading tests, and its expression is:

[0035] Where v is the vehicle speed, following a normal or Weibull distribution; w is the axle load, with the extreme value type I distribution used to fit the axle load spectrum distribution of the rear of heavy-duty trucks; k, α, and β are empirical coefficients obtained by fitting experimental data using the nonlinear least squares method; η c The coupling efficiency between the road surface and the piezoelectric unit is α, ranging from 0.6 to 0.9. lane Lane coverage.

[0036] Let p be the probability of the i-th car model appearing. i Its typical axle load is w i Typical vehicle speed is v i Then the expected value and variance of the piezoelectric energy of a single vehicle are respectively:

[0037]

[0038] Within the time interval Δt, the number of vehicles arriving, N, follows a Poisson distribution with a mean of λΔt. According to the composite distribution theory, the total piezoelectric output energy E within the time interval... total The expected value and variance are respectively:

[0039]

[0040] Piezoelectric output power P p =E total / Δt, its expectation and variance can be derived accordingly. When the number of vehicles within Δt is large, according to the central limit theorem, P p It approximately follows a normal distribution, that is:

[0041] Thus, the probability distribution function of the piezoelectric output power is obtained. .

[0042] S202: Establish the probability distribution function of photovoltaic output power Based on the sunshine probability in meteorological data, a probability distribution function for photovoltaic output power is established. First, historical daily sunshine intensity data for at least 10 years is extracted from meteorological data, and a probability distribution of sunshine intensity G is fitted. This invention uses a Beta distribution to describe the probability density function of sunshine intensity:

[0043] Where g is the normalized illumination intensity (the difference between the actual illumination intensity and the illumination intensity under standard test conditions G). STC =1000W / m 2 The ratio of α and β is a distribution parameter that varies with the seasons and can be obtained by fitting historical data through maximum likelihood estimation.

[0044] Introducing dynamic occlusion coefficient β s (t) Correction for photovoltaic output. This coefficient is calculated in real time based on road direction, height of surrounding features, and solar azimuth angle, and ranges from 0 to 1. The actual output power P of the photovoltaic module. v The relationship with effective light intensity is as follows:

[0045] Among them, P STC For peak power under standard test conditions, G eff =g·(1-β s (t) represents the effective normalized illumination intensity after shading correction, γ is the temperature coefficient (typically -0.36% / ℃), and T is the battery temperature (which can be ambient temperature +25℃). To simplify the model, temperature changes can be ignored or treated as deterministic corrections.

[0046] Because of P v With G eff The relationship is linear, and the probability density function of photovoltaic output power can be obtained through variable transformation. Let the conversion efficiency η be... v =P STC ·[1+γ(T-25)], then P v =η v ·G eff Due to β s (t) is a constant at a given time, and G is a constant. eff P is linearly proportional to g, therefore v The probability density function is:

[0047] Substituting into the Beta distribution expression, we obtain the probability distribution function of photovoltaic output power.

[0048] S203: Construct a system total power fluctuation model for the combined output of piezoelectric and photovoltaic power. Based on the established probability distribution functions of piezoelectric and photovoltaic output power, a system total power fluctuation model is constructed under the combined output of the two. Assuming that the piezoelectric and photovoltaic outputs are independent (and that the randomness of traffic flow and the randomness of sunlight have no direct causal relationship), the system total power P... t The sum of the two:

[0049] Due to independence, the probability density function of the total system power is the convolution of the piezoelectric power output probability density function and the photovoltaic power output probability density function:

[0050] Two volatility metrics are defined to quantify the volatility characteristics of the system's total power: One is the output power variance, which measures the degree of dispersion of the total power around the mean and reflects the fluctuation range:

[0051] The second is the flexibility deficit F. d It is used to measure the risk of insufficient flexibility in the system's supply and demand balance, taking into account both power shortage risk and power curtailment risk:

[0052] Among them, L d For the load power requirement, B c This represents the absorbable capacity of the energy storage system under its current state (i.e., the remaining charging capacity of the energy storage). The first term represents the expected shortfall in unmet load demand, and the second term represents the expected curtailment of power exceeding load demand that the energy storage system cannot absorb. d The smaller the size, the better the system flexibility.

[0053] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0054] S301. Definition of Decision Variables Suppose that the key configuration parameters to be optimized include the following three decision variables: piezoelectric unit layout density The number of piezoelectric units deployed per unit length of road segment, expressed in units / km, is expressed as follows:

[0055] in This represents the total number of piezoelectric units. The length of the road segment is (km).

[0056] Photovoltaic installed capacity The total peak power of a photovoltaic module, expressed in kW, is as follows:

[0057] in For the first Usable area (such as roadside slopes, sound barriers, building roofs, parking lot awnings, etc.). The corresponding laying efficiency (taking into account factors such as tilt angle, orientation, spacing, and occlusion). The light intensity is under standard test conditions.

[0058] Energy storage configuration capacity: including power capacity (kW) and energy capacity (kWh), both together determine the peak shaving and valley filling and emergency power supply capabilities of energy storage.

[0059] The decision variable vector is denoted as .

[0060] S302. Optimize the objective function Two conflicting objective functions are defined to characterize economic efficiency and power supply volatility, respectively.

[0061] Objective 1: Minimize annualized overall cost Annualized comprehensive cost The annual value of the initial investment, annual operation and maintenance costs, and annual cost of electricity purchased from external sources (if the system is connected to the grid) are expressed as follows:

[0062] in: This represents the total initial investment, including the cost of purchasing and installing piezoelectric units, photovoltaic modules, and energy storage systems.

[0063] , , These are the unit price of piezoelectric unit (RMB / unit), unit price of photovoltaic capacity (RMB / kW), and unit price of energy storage capacity (RMB / kWh).

[0064] This is the capital recovery coefficient. ,in The discount rate is... The system's design life (in years).

[0065] Annual maintenance costs are typically estimated as a percentage of the initial investment, such as... , Take 1% to 3%.

[0066] This represents the annual cost of electricity purchased from outside the grid, and applies only to grid-connected systems. For off-grid systems, this value is zero, but a self-consistency constraint must be met.

[0067] Objective 2: Minimize output power fluctuations Output power fluctuation is based on the total system power. variance Quantization is performed; the smaller the value, the more stable the power output. The objective function is:

[0068] in This represents the energy storage discharge power (positive value). The energy storage charging power is considered negative. In practical optimization, a time-series simulation of 8760 hours per year is typically used to calculate the sample variance of the net power series as a volatility indicator.

[0069] S303. Constraints This method comprehensively considers various constraints such as road engineering, economic efficiency, and power supply reliability, as detailed below: The deployment area of ​​piezoelectric units is limited by the excavable area of ​​the road surface and the lane coverage rate due to road land constraints.

[0070] in The floor space occupied by a single piezoelectric unit, This represents the upper limit of the total area where piezoelectric units can be installed in a road section. This refers to lane coverage. The area for photovoltaic module deployment is limited by available space due to roadside slopes, sound barriers, service area building roofs, parking lot awnings, etc.

[0071] in For the first The area actually used for photovoltaic deployment in this type of region. This represents the total usable area of ​​this type of region.

[0072] Investment budget constraints: The total initial investment must not exceed the given investment budget ceiling. :

[0073] Self-consistency rate constraint (required for off-grid systems, optional for grid-connected systems) Self-consistency rate Defined as the proportion of the system's annual power generation that is directly supplied to the load and is not curtailed, it is typically required to be no less than 95%.

[0074] in This represents the total annual power generation. To the extent of wasted electricity, This represents the total load demand.

[0075] Curtailment rate constraint The ratio of abandoned electricity to total power generation is typically required to not exceed 12% to prevent resource waste.

[0076] The energy capacity of the emergency backup constrained energy storage system must meet the power supply requirements of critical loads in emergency situations:

[0077] in For the power requirements of critical loads (such as emergency lighting, drainage pumps, and monitoring equipment), This is for emergency backup time. For general roads, The time limit is 2 hours; for highway tunnels or special important road sections, it can be extended to 12 hours.

[0078] When the voltage fluctuation constraint system is running, the load-side voltage fluctuation rate The tolerance threshold of the electrical equipment must not be exceeded (usually ±5% / min):

[0079] in The value is determined based on the voltage sensitivity of road electrical equipment (such as LED lighting and surveillance cameras), and is generally taken as 0.05.

[0080] S304. Multi-objective optimization solution methods Since the two objective functions mentioned above have different dimensions and are mutually restrictive (increasing energy storage can smooth out fluctuations but will increase costs), this method adopts a non-dominated sorting genetic algorithm (NSGA) with an elitist strategy. II) Solve for the Pareto optimal solution set. The specific process is as follows: Population initialization: Randomly generate an initial population of size 200, with each individual corresponding to a set of decision variables. And ensure that the hard constraints among the above constraints (such as land use and investment limits) are met.

[0081] Fitness evaluation: For each individual in the population, the annualized comprehensive cost is calculated through a time-series simulation of 8760 hours per year (step size 15 minutes or 1 hour). Output power variance Verify each soft constraint (self-consistency rate, curtailment rate, emergency reserve, voltage fluctuation) and impose penalties on individuals that violate the constraints.

[0082] Non-dominated ordination: The population is divided into different fronts according to Pareto dominance, and the crowding distance of each individual is calculated to maintain the diversity of solutions.

[0083] Genetic operations: The offspring population is generated using tournament selection, simulated binary crossover (crossover probability 0.85), and polynomial mutation (mutation probability 0.15).

[0084] Elite retention: The parent and offspring generations are merged, and a rapid non-dominated sorting and crowding comparison are performed to select the top 200 individuals as the new generation population.

[0085] Termination criteria: The iteration terminates when the maximum number of generations is 500, or when the rate of change of the hypervolume index is less than 1% for 50 consecutive generations.

[0086] The final Pareto front solution set contains multiple non-dominated candidate capacity configurations, each achieving an optimal trade-off between economy and volatility.

[0087] In one embodiment of the present invention, based on step S4, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0088] S401. Typical Working Condition Classification and Probability Determination Based on historical meteorological and traffic flow data, various typical operating conditions and their probabilities of occurrence are determined. Let there be a total of... A typical working condition, the first The probability of this working condition occurring is ,satisfy In this invention, typical operating conditions include at least the following three categories: Normal weather conditions: Regular weather conditions where photovoltaic power output is close to the historical average level, with a probability denoted as [probability value missing]. ; Continuous rainy weather condition: A prolonged period of rainy weather that causes a significant drop in photovoltaic power output, with a probability denoted as [missing information]. ; Holiday congestion scenario: An extreme traffic scenario in which a surge in traffic volume leads to increased piezoelectric output and increased charging load, with a probability denoted as . .

[0089] The above probabilities are determined based on statistical analysis of at least 10 years of historical data from local weather stations and traffic detectors, for example... , , .

[0090] S402. Joint Probabilistic Scenario Generation Multiple joint probability scenarios are generated using the Monte Carlo method, with each scenario corresponding to one simulated sampling. The specific process is as follows: Operating condition type sampling: Based on the probability of occurrence of each working condition The type of operating condition in this simulation was determined by random sampling using the roulette wheel method. Define the cumulative probability function:

[0091] Generate compliance Uniformly distributed random numbers Select the one that satisfies The smallest This is the operating condition type used in this simulation.

[0092] Generate lighting scene sequence: When sampling normal weather conditions or continuous rainy conditions, a first-order Markov chain model is used to generate the corresponding lighting scene sequence. The lighting conditions are divided into three categories: sunny (…). ),partly cloudy( ), rainy A first-order Markov chain is defined by its state transition probability matrix. describe:

[0093] in ,satisfy The transition probability matrix is ​​obtained based on historical meteorological data, and different transition probabilities can be applied to different operating conditions, such as continuous rainy conditions. (From cloudy / rainy to cloudy / rainy) significantly increased.

[0094] Given an initial state (e.g., the historical daily average state), the next state is sampled sequentially according to a time step (e.g., 15 minutes), generating a line of length [missing information]. The illumination state sequence (e.g., the number of time steps corresponding to 8760 hours throughout the year). Then, based on the illumination state at each moment, combined with the illumination intensity probability distribution under that state (e.g., a high-mean Beta distribution corresponding to a sunny day), the illumination intensity value at that moment is randomly sampled. Introducing a dynamic occlusion coefficient After that, the effective light intensity is The photovoltaic output power is then calculated using a photovoltaic output model. .

[0095] Generate traffic flow scene sequence: When sampling normal weather conditions or holiday congestion conditions, a non-homogeneous Poisson process is used to generate traffic flow scene sequences. Vehicle arrival rate. For time The function, fitted based on historical traffic flow data, can be represented using a Fourier series:

[0096] in For average arrival rate, This represents the harmonic order (typically 3-6). For congested holiday conditions, a peak amplification factor is superimposed on the arrival rate function. :

[0097] Determined based on the ratio of historical holiday traffic peak to weekday average, typical range of values. .

[0098] In length Number of vehicles arriving within the time step Obtain the parameter as The piezoelectric output power follows a Poisson distribution. Based on the piezoelectric output power probability model established in step S201, the piezoelectric output power during that time period is sampled from the distribution of the number of arriving vehicles and the piezoelectric energy of a single vehicle. Meanwhile, in response to traffic congestion during holidays, the power output of charging stations in the load demand needs to be increased accordingly, with the amplification factor being consistent with the peak traffic flow factor.

[0099] Combined to form a joint probability scenario: The complete time series generated under the same operating condition (including lighting scene sequence and traffic flow scene sequence) are aligned according to the time step to form a joint probabilistic scene. Let the total simulation time length be... (e.g., 8760 hours, step size) minutes, then the total number of steps is Then, a joint probability scenario can be represented as:

[0100] in The required power for the load can be determined according to the operating strategy of the load type (lighting, monitoring, charging piles, etc.). During holidays and congested conditions, the load on the charging piles is increased proportionally to the arrival rate.

[0101] Repeat sampling: Repeat the above steps to generate a total of A joint probability scenario (usually) These constitute the scenario pool for the Monte Carlo simulation. The frequency of occurrence of each scenario will approximate a preset probability distribution of operating conditions.

[0102] S403. Calculation of Flexibility Supply Guarantee Rate For each candidate capacity configuration scheme (a set of decision variables in the Pareto front solution set output from step S3) The system is simulated under various joint probability scenarios, and the operating strategy is as follows: At each time step, priority is given to using piezoelectric and photovoltaic power to meet load demands. ; If the power generation exceeds the load demand, the excess power will be used to charge the energy storage; if the energy storage is full, power will be wasted. If the power generation is insufficient to meet the load demand, the energy storage will discharge to supplement it. If the energy storage capacity is insufficient, the power will be purchased from the grid (grid-connected system) or non-critical loads will be disconnected (off-grid system).

[0103] During the simulation, record whether the following two conditions are met simultaneously at each time step: Supply and demand balance: load power Fully satisfied (critical loads must be satisfied, flexible loads may allow short-term derating). Voltage fluctuation constraint: Voltage fluctuation rate shall not exceed a preset threshold. (Usually taken as 5% / min).

[0104] Suppose that in a joint probability scenario, the number of time steps to satisfy the above two conditions is... Then the scene satisfaction in this scenario is .

[0105] For all After simulating each scenario, the flexibility supply guarantee rate is calculated using scenario satisfaction. Defined as meeting the satisfaction threshold (e.g.) The proportion of scenes with the following characteristics:

[0106] in For indicator functions, This is the scene satisfaction threshold (usually 0.95 or 0.99). In engineering, it can also be directly defined. The average satisfaction rate across all time steps in all scenarios:

[0107] Both definitions are acceptable; this embodiment uses the latter to simplify calculations.

[0108] S404. Configuration Scheme Filtering Preset reliability threshold , usually requires For each candidate capacity configuration, its flexibility supply guarantee rate is calculated, and those below the threshold are eliminated. The remaining configurations are then selected as the final configurations that meet the reliability requirements of road conditions.

[0109] This invention provides the following application examples: Example 1: This embodiment applies the capacity configuration method provided by the present invention to the design of a power supply system for traffic monitoring and information screens along urban main roads. The complete implementation steps S1 to S5 are as follows: Step S1 involves collecting road scene parameters and extracting features. Traffic flow detector data for the road segment over one year is collected, including traffic volume by vehicle type, time-averaged vehicle speed, and vehicle weight distribution. The data sample covers the entire year to reflect seasonal variations. Axle load spectrum distribution is obtained through the WIM system, covering typical axle load ranges for different vehicle types such as passenger cars, buses, and trucks. Vehicle speed distribution characteristics are obtained through radar detectors, reflecting speed variation patterns under free-flow and congested conditions. Lane coverage is determined based on lane function division; in this embodiment, piezoelectric units are deployed in all lanes. Local meteorological station data for the same period is collected, including daily solar intensity, ambient temperature, cloud cover, and rainfall probability, with a data period of at least 10 years to ensure statistical significance. A probability distribution of solar intensity is fitted based on historical solar data, using a Beta distribution, with distribution parameters varying seasonally. A dynamic shading coefficient database is established through on-site measurements or a BIM model, and the shading coefficient is calculated in real-time based on road orientation, surrounding building height, and solar azimuth. Collect load data, including the rated power and operating strategies of devices such as surveillance cameras, information display screens, and roadside units. Classify loads into rigid loads and flexible loads according to power supply reliability requirements. Rigid loads are required to have a power interruption time of no more than 2 seconds, while flexible loads are allowed to be derated proportionally or interrupted for a short time.

[0110] Output data of the piezoelectric unit under different load amplitudes and frequencies were obtained through indoor MTS loading tests. The empirical coefficients in the single-vehicle piezoelectric energy model were fitted using the nonlinear least squares method to establish a quantitative relationship between piezoelectric output and vehicle speed and axle load. Based on the IV curve under standard photovoltaic module test conditions, and combined with light intensity and temperature correction coefficients, a photovoltaic output model was established, and its accuracy was verified through field measurement data.

[0111] Step S2 constructs a road heterogeneous energy flow fluctuation model. Based on the randomness of traffic flow, a probability distribution function for piezoelectric output power is established. The day is divided into 96 time periods, and the KS test is used to verify whether the traffic arrival process in each time period conforms to a non-homogeneous Poisson process. A vehicle arrival rate function is fitted, and the expected and variance of piezoelectric output power are calculated by combining the expected and variance of piezoelectric energy per vehicle. Based on a Beta distribution fitting of light intensity, the probability density function of photovoltaic output power is obtained through variable transformation. A system total power fluctuation model under combined piezoelectric and photovoltaic output is constructed, defining output power variance and flexibility deficit as fluctuation indicators, where the flexibility deficit comprehensively considers the risks of power shortage and power curtailment.

[0112] Step S3 optimizes capacity configuration under road scenario constraints. Decision variables include piezoelectric unit deployment density, photovoltaic (PV) capacity, and energy storage capacity. Piezoelectric unit deployment density is determined based on the available road surface area and lane coverage. PV capacity is determined based on available area and installation efficiency of roadside slopes, sound barriers, and building rooftops. Energy storage capacity considers peak shaving and valley filling needs as well as emergency backup requirements. With the dual objectives of minimizing annualized comprehensive cost and minimizing output power fluctuations, and considering constraints such as road land use, investment budget, self-consistency rate, curtailment rate, emergency backup, and voltage fluctuations, a non-dominated sorting genetic algorithm with an elitist strategy is used to solve for the Pareto front. Algorithm parameters are set according to the problem size; population size, crossover probability, mutation probability, and maximum number of iterations are determined through pre-experiments. The termination criterion is that the rate of change of the hypervolume index is less than a set threshold for multiple consecutive generations. After obtaining the Pareto front, feasible solutions satisfying the probability constraints are selected based on the confidence probability.

[0113] Step S4 is executed to assess road condition reliability. The flexibility supply guarantee rate is introduced as an evaluation metric, defined as the probability that the system can meet flexibility requirements over a long period. Monte Carlo simulations are used to generate joint probability scenarios of traffic flow and illumination. The scenario proportions are determined based on historical data from local weather stations, including extreme conditions such as normal weather, continuous rain, and holiday congestion. The illumination scenario is generated using a first-order Markov chain model, with states categorized as sunny, cloudy, and rainy. The state transition probability matrix is ​​obtained from historical meteorological data. The traffic flow scenario is simulated using a non-homogeneous Poisson process, with the traffic flow in the holiday congestion scenario amplified according to historical peak values. The system is simulated in minute-level steps under each scenario, and the percentage of time when supply and demand are balanced and voltage fluctuation does not exceed a threshold is used as an estimate of the flexibility supply guarantee rate.

[0114] Step S5 outputs the road scenario configuration scheme. Based on actual engineering needs, a scheme that meets the reliability threshold is selected from the Pareto solution set, generating a heterogeneous energy configuration list for this road section, including recommended values ​​for piezoelectric unit deployment density, photovoltaic installed capacity, and energy storage configuration capacity. After the configuration list is submitted to the design unit, it is transformed into a specific engineering implementation plan through construction drawing design, including the piezoelectric unit deployment location and spacing, photovoltaic module installation method and orientation, energy storage system layout location and safety distance, etc.

[0115] The system apparatus for implementing the above method is as described in claim 8, and its specific structure and workflow have been embodied in the method embodiments, and will not be repeated here.

[0116] Example 2: This embodiment applies the method of the present invention to the configuration of off-grid power supply systems in highway service areas. Service areas are far from the power grid and require off-grid power supply systems to meet the loads of convenience stores, charging stations, sewage treatment facilities, etc. The specific implementation follows the process of: step S1 data acquisition, step S2 fluctuation model construction, step S3 optimization solution, step S4 reliability assessment, and step S5 solution output. Step S1 involves collecting road scene parameters and extracting features. This includes collecting traffic flow data from the service area's entrance and exit lanes, including the average daily number of vehicles entering and exiting, vehicle type distribution, dwell time distribution, and charging demand characteristics; collecting local weather station data, focusing on the probability distribution of light intensity and the probability of continuous rainy weather; and collecting load data, including the power requirements of equipment such as convenience store lighting and refrigeration equipment, DC fast charging piles, sewage treatment facilities, and site lighting. Based on power supply reliability requirements, sewage treatment facilities and emergency lighting are classified as rigid loads, while charging piles, convenience store lighting, and site lighting are classified as flexible loads. Piezoelectric units are deployed in the service area's entrance and exit lanes, photovoltaic modules utilize the service area parking lot awnings and building rooftop space, and the energy storage system is deployed in a safe area at the edge of the service area.

[0117] The methods for calibrating the piezoelectric energy model and establishing the photovoltaic output model are the same as in Example 1, but the influence of the slow speed and long dwell time of vehicles entering and leaving the service area on the piezoelectric output needs to be considered, as well as the constraints of the awning structure on the installation angle and shading coefficient of the photovoltaic modules.

[0118] Step S2 constructs a heterogeneous energy flow fluctuation model. A probability distribution function for piezoelectric output power is established based on the randomness of traffic flow in the service area, considering the arrival patterns and dwell time distribution of vehicles entering and leaving the service area. A probability distribution function for photovoltaic output power is established based on the probability distribution of light intensity, considering the dynamic shading of the photovoltaic modules by the surrounding environment of the service area. The variance and flexibility deficit of the combined piezoelectric and photovoltaic output are calculated to provide a basis for subsequent optimization.

[0119] Step S3 is executed to optimize capacity configuration. Decision variables include the density of piezoelectric units in service area access lanes, the photovoltaic capacity of parking lot awnings, and the capacity of the energy storage system. Constraints include available land area in the service area, investment budget limitations, off-grid system self-consistency requirements, curtailment rate limitations, emergency backup requirements, and voltage fluctuation constraints. With the dual objectives of minimizing annualized comprehensive cost and minimizing output power fluctuation, a multi-objective optimization algorithm is used to solve for the Pareto front, and feasible solutions are selected based on confidence probabilities.

[0120] Step S4 is executed to perform a reliability assessment. A combined probabilistic scenario of traffic flow and sunlight is generated using Monte Carlo simulation. Extreme conditions are considered, such as continuous rainy weather leading to a significant drop in photovoltaic output, and holiday congestion due to a surge in traffic and increased charging demand. The flexibility and supply guarantee rate of each configuration is calculated. Particular attention is paid to whether the system can guarantee power supply to rigid loads such as sewage treatment and emergency lighting under extreme scenarios of continuous rainy weather combined with peak charging periods.

[0121] Step S5 outputs the configuration scheme. Based on the actual engineering needs, select the scheme that meets the reliability threshold and generate a configuration list, including recommended values ​​for the piezoelectric unit deployment density of service area access lanes, photovoltaic installed capacity of parking lot sunshades, and energy storage system capacity, and provide performance indicators such as emergency backup time for critical loads, expected annual power generation, and self-sufficiency rate.

[0122] Example 3: This embodiment applies the method of the present invention to the power supply configuration of a lighting system at the entrance section of a long tunnel. The tunnel entrance section requires the installation of enhanced lighting, transitional lighting, basic lighting, and emergency lighting. Emergency lighting is a rigid load, while enhanced and transitional lighting are flexible loads, adjustable according to traffic volume and external tunnel brightness. The specific implementation is as follows: Step S1 involves collecting road scene parameters and extracting features. Traffic flow data at the tunnel entrance is collected, including average daily traffic volume, vehicle speed distribution, and vehicle type ratio, with vehicle speeds at the deceleration section before entering the tunnel approximately 20-40 km / h. Meteorological data at the tunnel entrance is also collected, including light intensity and surrounding mountain shading. Lighting load data is collected, and the power density and total load for different lighting sections are calculated according to the "Detailed Design Specifications for Highway Tunnel Lighting." Piezoelectric units are deployed on the road surface at the tunnel entrance, utilizing the load energy during vehicle deceleration. Photovoltaic modules utilize the tunnel entrance slope and shading space, and the impact of dynamic mountain shading on photovoltaic output must be considered.

[0123] When calibrating the piezoelectric energy model, special consideration must be given to the impact of vehicle deceleration on speed distribution, as well as the influence of the road surface structure within the tunnel on the coupling efficiency of the piezoelectric units. When establishing the photovoltaic output model, a dynamic shading coefficient calculation model needs to be developed based on the tunnel alignment and the height of the surrounding mountains.

[0124] Step S2 constructs a heterogeneous energy flow fluctuation model. A probability distribution function for piezoelectric output power is established based on the randomness of traffic flow at the tunnel entrance, considering the Poisson process of vehicle arrival and the deceleration distribution of vehicle speed. A probability distribution function for photovoltaic output power is established based on light intensity and dynamic shading, considering the variation of mountain shading with the solar azimuth angle. The variance and flexibility deficit of the combined piezoelectric and photovoltaic output are calculated to provide a basis for capacity allocation.

[0125] Step S3 is executed to optimize capacity configuration. Decision variables include the piezoelectric unit deployment density at the tunnel entrance, the photovoltaic capacity of the tunnel entrance slope and sunshade, and the energy storage system capacity. Constraints include available deployment area at the tunnel entrance, investment budget limitations, tunnel lighting self-sufficiency requirements, curtailment rate limitations, emergency backup requirements, and voltage fluctuation constraints. With minimizing annualized comprehensive cost and output power fluctuation as dual objectives, a multi-objective optimization algorithm is used to solve for the Pareto front.

[0126] Step S4 is executed to perform a reliability assessment. A joint probability scenario of traffic flow and illumination is generated using Monte Carlo simulation, considering extreme conditions such as heavy rain and holiday congestion, to calculate the flexibility and supply guarantee rate of each configuration. Particular attention is paid to whether the system can guarantee power supply to rigid loads such as emergency lighting and drainage pumps in the event of a mains power outage or insufficient photovoltaic output due to heavy rain.

[0127] Step S5 outputs the configuration scheme. Based on the importance and reliability requirements of tunnel lighting, select the scheme that meets the reliability threshold, generate a configuration list, including recommended values ​​for piezoelectric unit deployment density at the tunnel entrance, photovoltaic installed capacity at the entrance, and energy storage capacity, and provide performance indicators such as emergency lighting backup time, expected annual power generation, and reduction ratio of power supply fluctuations.

[0128] Please refer to Figure 4 The present invention also provides a capacity configuration device for a road heterogeneous energy complementary power generation system, including a data acquisition module, a model building module, an optimization solution module, a reliability assessment module, and a scheme output module.

[0129] The data acquisition module consists of a dynamic weighing system, a weather station interface, and a smart meter. It is used to collect traffic flow data, meteorological data, and load data in real time or offline, and then store the multi-source data in a database after uniform formatting.

[0130] The model building module is a computing unit embedded with a piezoelectric power output model, a photovoltaic power output model, and a wave model. It calibrates the model parameters based on the data provided by the data acquisition module and generates a heterogeneous energy flow wave model.

[0131] The optimization solution module is a server or industrial control computer equipped with the NSGA-II multi-objective optimization algorithm. It receives the fluctuation model output by the model building module and solves the Pareto front under the set decision variable range and constraints.

[0132] The reliability assessment module is a computer unit that runs a Monte Carlo simulation program, generates a combined traffic flow-lighting probability scenario, and calculates the flexibility supply guarantee rate of each configuration scheme.

[0133] The solution output module is a human-computer interaction interface or report generation unit that displays a list of configurations that meet the reliability threshold.

[0134] The above modules are connected in sequence: the output of the data acquisition module is connected to the input of the model building module; the output of the model building module is connected to the input of the optimization solution module; the output of the optimization solution module is connected to the input of the reliability assessment module; and the output of the reliability assessment module is connected to the input of the scheme output module.

[0135] Compared to ordinary photovoltaic-storage microgrids, this invention has the following road-specific attributes: Road-driven power characteristics: piezoelectric output is driven by traffic flow, and its randomness exhibits traffic engineering patterns such as peak hours and holiday effects; the piezoelectric pulse characteristics require consideration of the capacity requirements of energy storage and inverters due to instantaneous power surges; Linear constraints on spatial resources: road space is linear and strip-shaped, with the deployment length proportional to the road segment length and the width limited by the road cross-section; the deployment of piezoelectric units must consider lane coverage, pavement structure layer thickness, and the fatigue impact of traffic loads on piezoelectric materials; Traffic-related load types: tunnel lighting needs to be dimmed according to traffic volume, and the load of charging piles in service areas is related to the number of vehicles entering; capacity configuration requires establishing a "vehicle flow-photovoltaic-load" coupling model; Differentiated reliability requirements: emergency loads require 99.99% reliability, while conventional lighting can accept 95%; adopting layered reliability constraints can avoid investment waste.

[0136] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. A capacity configuration method for a road-based heterogeneous energy complementary power generation system, characterized in that, include: Obtain traffic flow data and weather data for the target road segment; Based on the traffic flow data and meteorological data, a heterogeneous energy flow fluctuation model reflecting the joint fluctuation characteristics of piezoelectric power output and photovoltaic power output is constructed. Based on the output of the model, the output power variance and flexibility deficit are determined as indicators for evaluating power generation fluctuation. With the dual objectives of minimizing annualized comprehensive cost and minimizing the aforementioned indicators, and with piezoelectric unit deployment density, photovoltaic installed capacity, and energy storage configuration capacity as decision variables, a multi-objective optimization algorithm is used to solve for the candidate capacity configuration parameters under constraints including road land use constraints and investment budget constraints. A joint probability scenario of traffic flow and illumination is generated through Monte Carlo simulation. The reliability of each candidate capacity configuration parameter is evaluated through the joint probability scenario, and the capacity configuration parameter that meets the preset reliability threshold is output. The capacity configuration parameters include the piezoelectric unit deployment density, the photovoltaic installed capacity, and the energy storage configuration capacity.

2. The method according to claim 1, characterized in that, Obtain traffic flow data and meteorological data for the target road segment, including: Obtain the axle load spectrum distribution, vehicle speed distribution, lane coverage, and vehicle arrival rate from the traffic flow data; Obtain the probability distribution of light intensity, effective sunshine duration, and dynamic shading coefficient from the meteorological data.

3. The method according to claim 2, characterized in that, Acquiring traffic flow and weather data for the target road segment also includes: The axle load spectrum distribution, the vehicle speed distribution, and the vehicle arrival rate are obtained by a dynamic weighing system or traffic flow detectors deployed on the target road section. The lane coverage rate is calculated and determined based on the total number of lanes in the target road segment and the planned number of lanes for the piezoelectric unit deployment; The light intensity probability distribution and the effective sunshine duration are obtained by collecting meteorological station or satellite remote sensing data near the target road section; The dynamic shading coefficient is obtained by measuring the road direction, the height of surrounding features, and the solar azimuth angle on site, combined with pre-calculation using solar radiation analysis software or by establishing a shading coefficient database using a three-dimensional model.

4. The method according to claim 1, characterized in that, Based on the aforementioned traffic flow data and meteorological data, a heterogeneous energy flow fluctuation model reflecting the joint fluctuation characteristics of piezoelectric and photovoltaic power output is constructed, including: Based on the randomness of traffic flow data, a probability distribution function for piezoelectric output power is established; Based on the probability of sunshine in meteorological data, a probability distribution function for photovoltaic output power is established; Based on the probability distribution function of the piezoelectric output power and the probability distribution function of the photovoltaic output power, a system total power fluctuation model of the combined piezoelectric and photovoltaic output power is constructed.

5. The method according to claim 4, characterized in that, Based on the randomness of traffic flow data, a probability distribution function for piezoelectric output power is established, including: Axle load spectrum distribution, vehicle speed distribution, lane coverage, and vehicle arrival rate are extracted from the traffic flow data. Based on the empirical relationship between single vehicle piezoelectric energy and axle load and vehicle speed calibrated by indoor loading tests, and combined with the lane coverage rate, a single vehicle piezoelectric energy model is established. Based on the vehicle arrival rate, the probability distribution of the number of vehicles arriving within a time period is determined. Combined with the single vehicle piezoelectric energy model, the expectation and variance of the piezoelectric output power within the time period are calculated, thereby obtaining the probability distribution function of the piezoelectric output power.

6. The method according to claim 4, characterized in that, Based on the sunshine probability in meteorological data, a probability distribution function for photovoltaic output power is established, including: Historical light intensity data are extracted from the meteorological data, and a probability distribution model of the original light intensity is obtained by fitting. The probability distribution model uses a Beta distribution to describe the distribution pattern of light intensity. A dynamic shading coefficient is extracted from the meteorological data. The dynamic shading coefficient is calculated based on the road direction, the height of surrounding features, and the solar azimuth angle, and is used to characterize the proportion of light intensity attenuated due to shading. The probability distribution model of the original illumination intensity is corrected using the dynamic shading coefficient to obtain the probability distribution of the effective illumination intensity after shading attenuation. Based on the output characteristics of photovoltaic modules, the probability distribution of the effective light intensity is converted into a probability distribution function of photovoltaic output power.

7. The method according to claim 4, characterized in that, Based on the probability distribution functions of the piezoelectric output power and the photovoltaic output power, a system total power fluctuation model for the combined piezoelectric and photovoltaic outputs is constructed, including: The probability distribution function of the piezoelectric output power and the probability distribution function of the photovoltaic output power within the same time period are convolved to obtain the probability distribution function of the total power. Based on the probability distribution function of the total power, the time-series fluctuation characteristics of the total power are calculated, and the time-series fluctuation characteristics include the output power variance; Based on the power balance relationship between total power and load demand, the flexibility deficit is calculated, which is used to characterize the degree of risk of not being able to meet load demand or absorb excess power.

8. The method according to claim 1, characterized in that, The constraints include: Road land constraints include: the piezoelectric unit deployment area is limited by the excavable area of ​​the target road section and the lane coverage rate, wherein the lane coverage rate is the ratio of the number of lanes where piezoelectric units are actually deployed to the total number of lanes; the photovoltaic module deployment area is limited by the available area of ​​the roadside slope, sound barrier, service area building roof or parking lot sunshade of the target road section. Investment budget constraints include: the sum of the purchase and installation costs of piezoelectric units, photovoltaic modules, and energy storage systems shall not exceed the preset investment budget ceiling.

9. The method according to claim 1, characterized in that, The candidate capacity configuration parameters are solved using a multi-objective optimization algorithm, including: A non-dominated sorting genetic algorithm with an elitist strategy is adopted as the multi-objective optimization algorithm. The decision variables are encoded to generate an initial population, including the piezoelectric unit deployment density, photovoltaic installed capacity, and energy storage configuration capacity. Based on the annual time-series simulation of the target road segment, the annualized comprehensive cost and output power fluctuation index of the configuration scheme corresponding to each individual in the population are calculated. Individuals in the population are sorted and selected according to non-dominated sorting and crowding distance, and offspring populations are generated through crossover and mutation operations; The population iteration is performed iteratively until the preset termination condition is met, and the final Pareto front solution set is output as a candidate capacity configuration parameter.

10. The method according to claim 1, characterized in that, A joint probabilistic scenario of traffic flow and illumination is generated through Monte Carlo simulation. The reliability of each candidate capacity configuration parameter is evaluated using this joint probabilistic scenario, and the capacity configuration parameters that meet a preset reliability threshold are output, including: Based on historical meteorological data and historical traffic flow data, various typical operating conditions and the probability of occurrence of each condition are determined. The typical operating conditions include at least normal weather conditions, continuous rainy conditions, and holiday congestion conditions. Based on the probability of occurrence of each working condition, the type of working condition for each simulation is determined by random sampling using the Monte Carlo method; When sampling normal weather conditions or continuous rainy conditions, a first-order Markov chain model is used to generate the corresponding lighting scene sequence; when sampling holiday congestion conditions, a non-homogeneous Poisson process is used and the traffic flow scene sequence is generated by scaling up according to the historical peak ratio. Align the lighting scene sequence and the traffic flow scene sequence generated under the same working condition according to the time step to form a joint probability scene; Repeat the above sampling and generation steps to obtain multiple joint probability scenarios; For each candidate capacity configuration parameter, the system is simulated under various joint probability scenarios. The proportion of time steps in which the system satisfies supply and demand balance and voltage fluctuation does not exceed a preset threshold is counted out of the total number of simulation steps. This proportion is used as the flexibility supply guarantee rate of the configuration scheme. The configuration scheme that achieves a flexible supply guarantee rate that meets the preset reliability threshold will be used as the final output capacity configuration parameter.

Images