Campus building photovoltaic potential assessment method based on parameterization analysis

Abstract

The embodiment of the invention provides a campus building photovoltaic potential assessment method based on parameterization analysis. The method comprises the steps that the potential generating capacity of a campus building roof photovoltaic system is obtained through analysis based on the annual irradiation intensity change rule; comparing and analyzing the potential generating capacity and the actual electrical load curve to determine a light abandoning rate distribution condition; constructing an energy storage system regulation and control model suitable for the campus scene based on the light abandoning rate distribution condition; and formulating an optimized energy storage scheduling strategy according to the regulation and control model so as to reduce the light abandoning rate of the photovoltaic system. Through the scheme of the embodiment of the invention, the problem of how to regulate and control the scheduling strategy of the energy storage system according to the change rule of the annual irradiation intensity so as to solve the problem that the light abandoning rate of the photovoltaic system is too high can be solved.

Description

Technical Field

[0001] This application relates to photovoltaic power generation technology, and specifically to a method for evaluating the photovoltaic potential of campus buildings based on parametric analysis. Background Art

[0002] The campus building photovoltaic potential assessment method based on parametric analysis is a technical means to evaluate the potential production capacity of photovoltaic power generation systems installed on the roofs and facades of campus buildings by quantitatively analyzing various influencing factors, such as building structure, lighting conditions, meteorological data and other parameters. This method can provide more accurate photovoltaic system layout recommendations, thereby maximizing power generation benefits. However, this method also faces some challenges, such as how to regulate the energy storage system scheduling strategy according to the changing pattern of irradiation intensity throughout the year. Since the power generation of the photovoltaic system is affected by the seasonal and diurnal variations in light intensity, a high curtailment rate may occur when the light resources are abundant (that is, the excess electricity cannot be effectively utilized). Therefore, it is necessary to combine energy storage technology and formulate a flexible scheduling strategy to reasonably absorb excess electricity and improve the overall economic benefits and energy utilization of the photovoltaic system. Summary of the Invention

[0003] In view of this, the embodiments of the present disclosure provide a method for evaluating the photovoltaic potential of campus buildings based on parametric analysis, which at least partially solves the problems existing in the prior art.

[0004] The photovoltaic potential assessment method for campus buildings based on parametric analysis includes:

[0005] The potential power generation of the photovoltaic system on the roof of campus buildings is obtained based on the analysis of the annual radiation intensity variation pattern;

[0006] Compare and analyze the potential power generation and actual power load curve to determine the distribution of curtailment rate;

[0007] Based on the distribution of the abandoned solar power rate, a storage system control model suitable for campus scenarios is constructed;

[0008] An optimized energy storage scheduling strategy is formulated based on the regulation model to reduce the curtailment rate of the photovoltaic system.

[0009] In a specific embodiment, the method of deriving the potential power generation of the campus building rooftop photovoltaic system based on the annual radiation intensity variation analysis further includes:

[0010] Model the daily average radiation intensity G based on different seasons;

[0011] The formula for calculating the total annual radiation energy Q is Q = \int_{0}^{365}G(t)\cdot dt, where t is the cumulative number of days in a year and G(t) is the average daily radiation intensity on day t;

[0012] A threshold T is set during the potential light abandonment period. When G(t)>T, the period is marked as a high irradiance output period.

[0013] A local control energy storage solution is constructed for the high irradiation output section.

[0014] In a specific embodiment, the following steps are also included:

[0015] Correct the potential power generation by analyzing the peak irradiation intensity P_m month by month throughout the year;

[0016] Calculate the monthly average peak adjustment ratio K_m = (P_max P_m) / P_max, where P_max represents the maximum peak radiation intensity of the whole year;

[0017] The formula C = \sum_{m=1}^{12}G_m\cdot K_m is introduced to calibrate the compensation of the total potential power generation for the whole year, where G_m represents the average potential power generation in the mth month;

[0018] The actual curtailment distribution curve is predicted based on the adjusted power generation.

[0019] In a specific embodiment, the comparative analysis of the potential power generation and the actual power load curve to determine the distribution of the abandoned solar power rate further includes:

[0020] The dynamic time-shift matching method is used to adjust the time deviation Δt of the two curves so that they can coincide more accurately.

[0021] Use the formula D_t = |\sum_{i=1}^N P_i L_i|, where P_i and L_i represent the values of photovoltaic power generation and actual power load at time step i respectively;

[0022] If the daily net difference in energy consumption D_t continuously exceeds the rated energy storage threshold C_th, this time period is marked as an overload risk period;

[0023] Design a time-sharing priority charging and discharging strategy for areas with high curtailment rates to reduce losses.

[0024] In a specific embodiment, constructing an energy storage system control model suitable for campus scenarios based on the abandoned solar power rate distribution further includes:

[0025] Set the maximum capacity S_MAX of the initial energy storage battery and record the current energy storage level H;

[0026] Establish a constraint function J = max(0, G_t L_t S_RATE\cdot R_H) to determine the real-time abandoned optical power, where S_RATE is the storage rate and R_H is the current charging rate.

[0027] If L_t+J>H+DIS_TH, the optimization mechanism will be triggered to reduce abandoned light;

[0028] Enable auxiliary energy supplement function during low sunlight period to alleviate power shortage problem.

[0029] In a specific embodiment, formulating an optimized energy storage scheduling strategy based on the control model to reduce the abandonment rate of the photovoltaic system further includes:

[0030] Plan a multi-stage energy storage strategy S_n based on the calendar periodicity characteristics and define the energy storage coefficient k_n for each stage;

[0031] Based on the condition E_{store}(n+1)=E_{current}\cdot e^{k_d\cdot d}, the remaining energy decreases with the waiting time, where k_d represents the natural loss rate and d represents the number of days of delay.

[0032] For key days (such as holidays), a tolerance factor for abandoned solar power is set up separately, and the optimized control rule is obtained by combining it with the real-time weather index W_r;

[0033] In the case of extreme radiation fluctuations, short-term excess usage of energy storage is allowed to ensure balanced and stable power supply.

[0034] In a specific embodiment, the method further comprises the following steps:

[0035] Taking into account the historical photovoltaic utilization rate R_h data and the predicted growth factor F_g in the next N years, the objective function G(x,y) = R_h*F_g^y / x is constructed, where x is the subjective optimization weight and y represents the incremental order;

[0036] The single-period energy storage dispatch cost B_t is calculated using the formula B_t=|G(t)P_d|\cdot\beta_p, where the β_p conversion factor depends on differences in electricity price structures.

[0037] Introducing the light abandonment penalty term P_l = η*(D_max / D_target 1)^2 into the overall scheduling algorithm to improve the optimization effect;

[0038] Dynamic threshold update rules are set for special meteorological conditions to maintain scheduling flexibility.

[0039] In a specific embodiment, the specific steps are as follows:

[0040] Extract the photovoltaic load ratio M_p of key nodes and the state vector X_s of the power station to locate the bottleneck link;

[0041] Optimize the configuration formula using the iterative correction method: ΔX_s(n) = α * [∂Λ(X_s) / ∂X_s], where α is the step size adjustment term and Λ is the energy efficiency evaluation target equation.

[0042] Find the optimal ratio value that satisfies the constraint conditions and minimizes the discarded power within the set of multiple possible solutions.

[0043] Implement simulation feedback testing to improve the energy storage distribution efficiency and prevent risks of redundancy or failure.

[0044] In a specific embodiment, particularly,

[0045] Propose a thermal balance evaluation sub-process for optimizing the heat dissipation effect under high temperature conditions during summer days.

[0046] Introduce an environmental temperature and humidity coupling effect estimation module: E_a = λ * exp((THTH_opt) / (δ_T)) * S_max, where TH is the temperature reference benchmark and λ is the heat dissipation efficiency parameter.

[0047] When the measured actual surface temperature TH_ex > TH_lim or the relative humidity exceeds the limit, immediately execute the refrigeration loss reduction measures to reduce the energy waste phenomenon caused by the performance degradation of the photovoltaic panel.

[0048] Improve the air circulation efficiency by reasonably designing the air duct direction to ensure that the overall operation stability is not overly disturbed by climate conditions.

[0049] In a specific embodiment, an additional step is added:

[0050] Utilize deep learning to挖掘 the long-term trend fitting ability in complex environments to form a decision basis matrix A_pred.

[0051] Define the photovoltaic load imbalance risk level scoring index: R_s = [∑|G_iL_j|i<j / MAX(|GiGavg||)|] * ξ, where ξ is the normalization scaling ratio factor.

[0052] Select the area with the highest probability through the activation layer as the action guidance for the next day and generate a preset action list L_optimal.

[0053] When a sudden drastic change in light occurs, promptly switch to the most adaptable emergency response mode to minimize the amplitude of temporary disturbance losses.

[0054] In a specific embodiment, it also involves supplementing detailed content:

[0055] Add an anomaly detection module to generate an intelligent response instruction set M_repair for the alarm information of equipment in abnormal working states.

[0056] By using the fault location probability derivation method, the possible problem parts and their impact ratio are found. Pi = f(i)*exp(di / c), where f is the deceleration factor and c is the threshold limit.

[0057] Integrate all diagnostic conclusions into a complete repair list for operation and maintenance engineers to review and process to improve maintenance efficiency;

[0058] Achieve all-round monitoring to ensure that each link always maintains the best operating state to avoid unnecessary energy overflow and waste.

[0059] The disclosed embodiments provide a method for evaluating the photovoltaic potential of campus buildings based on parametric analysis, including: determining the potential power generation of campus building rooftop photovoltaic systems based on an analysis of the year-round radiation intensity variation pattern; comparing the potential power generation with the actual power load curve to determine the distribution of the curtailment rate; constructing an energy storage system control model suitable for campus scenarios based on the curtailment rate distribution pattern; and formulating an optimized energy storage scheduling strategy based on the control model to reduce the curtailment rate of the photovoltaic system. The solution of the disclosed embodiments can solve the problem of how to regulate the energy storage system scheduling strategy based on the year-round radiation intensity variation pattern to address the problem of excessively high photovoltaic system curtailment rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the exemplary implementation methods of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the embodiments of the present disclosure and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 It is a flow chart of the photovoltaic potential assessment method for campus buildings based on parametric analysis;

[0062] Figure 2 This is a flow chart that further derives the potential power generation of the photovoltaic system on the roof of campus buildings based on the analysis of the annual radiation intensity variation pattern;

[0063] Figure 3 is a flow chart further comprising the steps of:;

[0064] Figure 4 It is a flow chart for further analyzing the potential power generation and actual power load curve to determine the distribution of curtailment rate;

[0065] Figure 5 This is a further flow chart of building an energy storage system control model suitable for campus scenarios based on the distribution of abandoned solar power rates;

[0066] Figure 6It is a further flow chart for formulating an optimized energy storage scheduling strategy based on the control model to reduce the abandonment rate of the photovoltaic system;

[0067] Figure 7 is a flow chart further comprising the steps of:

[0068] Figure 8 It also includes a flow chart of the following specific steps:

[0069] Figure 9 is, in particular, a flow chart of;

[0070] Figure 10 It is a flowchart with the additional steps:

[0071] Figure 11 It is a flowchart that also involves additional details: DETAILED DESCRIPTION

[0072] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present application. Therefore, the drawings and description are to be regarded as illustrative in nature and not restrictive.

[0073] Next, with reference to the accompanying figures, we will describe the present invention's parametric analysis-based method for assessing the photovoltaic potential of campus buildings. This method uses four key steps to scientifically and systematically assess the power generation potential of campus building photovoltaic systems. This method, combined with energy storage system optimization and regulation, addresses the issue of high solar curtailment. This not only improves energy efficiency but also lays the technical foundation for building sustainable, green campuses.

[0074] The first step is to derive the potential power generation of the photovoltaic systems on the rooftops of campus buildings based on an analysis of the annual variation in irradiance intensity. This step requires collecting the geographic location of the campus area and meteorological data for the area, including the distribution characteristics of solar radiation intensity throughout the year. By modeling historical radiation data and seasonal cycle patterns, it is possible to accurately simulate the power generation performance of photovoltaic power generation equipment at a monthly or even daily time resolution. For example, software tools such as MeteoDyn or PVSol can be used to input basic information such as latitude, longitude, and altitude to generate hourly solar resource status charts throughout the year. These radiation levels are then substituted into the standardized performance formula for a specific model of photovoltaic panel to estimate how much electricity each panel can theoretically output at different time periods.

[0075] The second step is to compare the potential power generation derived from the first step with the actual power load curve to determine the spatial and temporal distribution of curtailment. This requires a thorough understanding of the school's electricity consumption patterns, from the start of the morning teaching period to the load shedding after the library closes in the evening. Accurate data can be obtained through long-term operational monitoring records, or more comprehensive information can be obtained through on-site questionnaires and typical day measurements. In one example, assume that the average daily total power consumption of a university's main campus is close to 25,000 kWh, with peak usage occurring during the peak teaching period from noon to evening. If the calculation shows that the PV output exceeds 30% at noon but is completely absent at night, this clearly indicates that this excess power will be wasted without appropriate storage solutions, resulting in high curtailment.

[0076] Then, based on the detailed information provided in the above two steps, we proceed to the third step, which is to establish a framework for the energy storage system control system in the school context based on the already determined high rate of abandoned solar power. This is the most core and challenging part of the entire method, which involves the selection and balancing of multiple sub-objective functions such as economic cost minimization, carbon emission control, etc. Specifically, designers must comprehensively consider the characteristics of the energy storage battery type (such as lithium-ion or sodium-sulfur batteries), initial capacity specifications and factors affecting its cycle life to construct a mathematical programming model. For example, the net present value (NPV) may be selected as one of the evaluation indicators to introduce linear programming or mixed integer nonlinear optimization technology to solve the problem, and find a set of optimal parameters to ensure that the economic benefits are maximized while effectively reducing the waste ratio of abandoned solar power.

[0077] The final step involves the practical application of the previously established energy storage scheduling model and the development of a series of improvement measures to mitigate the current high risk of PV disuse. This step emphasizes the importance of feedback mechanisms—testing the effectiveness of different algorithms through simulations of various operating conditions, and continuously adjusting control rules until the desired performance standards are achieved. Assuming that after the previous stages, the team has reached some preliminary conclusions regarding which periods should prioritize charging and discharging, and how to set trigger thresholds. These settings will then be compiled into code and integrated into an automated platform to execute the command chain. If actual operational monitoring results indicate that on a sunny afternoon, partial shading may have led to an erroneous prediction of excessive storage, resulting in insufficient storage for students to study in the evening, the next day's operation plan will be adjusted immediately to increase daytime utilization and extend the evening discharge period.

[0078] In short, this method fully demonstrates how to tailor energy storage system scheduling strategies based on the year-round trajectory of solar irradiance, successfully addressing the high power loss problem on campus caused by the large fluctuations in photovoltaic system output. This multifaceted collaborative research approach has accumulated valuable experience and provided solid theoretical guidance for future universities to implement green and environmentally friendly new energy transformation projects.

[0079] Next, the present invention is further described to obtain the potential power generation of the photovoltaic system on the roof of the campus building based on the analysis of the change law of the radiation intensity throughout the year. In this method, multiple steps are included, which are used to model, quantify and optimize the potential power generation of the photovoltaic system on the roof of the campus building. First, modeling is performed based on the daily average radiation intensity G in different seasons. In this step, according to historical data, the year is divided into four seasons: spring, summer, autumn and winter, and the daily average radiation intensity function model of each time period is determined according to the characteristics of solar radiation in different seasons. This model can better reflect the change law of solar radiation under actual environmental conditions, thereby providing accurate basic input data for calculation.

[0080] Then, the total annual radiation energy Q is calculated by the integration formula. Specifically, the formula is Q = \int_{0}^{365}G(t)\cdot dt, where t represents the cumulative number of days in a year (ranging from 0 to 365), and G(t) is the average daily radiation intensity on day t, in kW / m 2 By cumulatively summing the average daily irradiance over time, we obtain the total annual photovoltaic energy Q, which is used to assess the total annual power generation potential. The formula must take into account the multiple influences of the Earth's revolution and rotation, as well as meteorological conditions, to ensure that the final energy estimate is close to actual conditions, thereby supporting better investment decisions and technology implementation.

[0081] At the same time, potential curtailment periods are defined, and high irradiance output segments are marked by a threshold value T. When the daily average irradiance intensity exceeds a certain preset standard, that is, G(t)>T, the current moment or stage is marked as a peak period with high light resources and which may cause the photovoltaic panel to generate electricity exceeding local consumption. For example, in a specific embodiment, the high irradiance intensity threshold value T can be set to 0.6kW / m 2 If G(t) is as high as 0.8kW / m between 12:00 noon and 2:00 pm on a certain day 2 , which indicates that this interval is a typical peak output period.

[0082] Finally, an energy storage solution is proposed for the peak output periods marked above to adjust the dynamic balance between campus photovoltaic power generation and electricity consumption. In order to achieve local regulation, it is necessary to design small energy storage facilities suitable for the scale of the campus, such as battery packs or supercapacitors. By introducing energy storage solutions, energy waste caused by excessive sunlight can be reduced, and the stability and sustainability of the campus energy system can be increased. Specifically, at noon in summer, energy storage equipment can store some excess electricity for use in subsequent non-sunlight periods, thereby improving the overall photovoltaic utilization rate. By combining the various contents and technical indicators in the above steps, the potential power generation of the photovoltaic system on the roof of campus buildings can be accurately evaluated.

[0083] Next, the present invention is described, which also includes the following steps:. Correcting the potential power generation by analyzing the peak irradiance intensity month by month throughout the year. This step means that when evaluating the photovoltaic potential of campus buildings, it is necessary to recalibrate the photovoltaic power generation that the building may generate based on the maximum peak irradiance intensity P_m of each month. This is to ensure the accuracy of the estimate, because different periods of the year (such as changes in winter and summer) will lead to significant differences in lighting conditions. For example, the peak irradiance intensity is usually lower in winter and higher in summer.

[0084] The formula for calculating the monthly average peak adjustment ratio is K_m = (P_max - P_m) / P_max. The parameters include P_max, which represents the maximum peak irradiance intensity for the entire year. It can theoretically be set to the highest measured value of the year, with the range of values determined by the specific geographic location and climate. Generally speaking, the highest point is reached during the months with the strongest sunlight, such as between July and September. The purpose of this formula is to quantify the impact of each month by comparing it to the annual peak. This formula is introduced because monthly irradiance fluctuations directly affect the actual energy conversion rate of solar panels, making calculations based solely on annual averages inaccurate. Therefore, a monthly adjustment is introduced to reflect a more detailed energy variation pattern.

[0085] Furthermore, the formula C = Σ(G_m·K_m) is used to compensate and calibrate the total power generation for the whole year. Among them, G_m means the average potential power generation in the mth month, and its unit should be consistent so that the total power generation for the year can be calculated cumulatively; the range of G_m will vary with the region and equipment capacity specifications, and the specific value can be referred to similar historical statistics. The core function of the formula is to combine the single-month radiation correction in the previous step with the corresponding power generation data to output a more appropriate overall estimated value. In one embodiment, assuming that the actual G_m of the photovoltaic system of a campus teaching building is higher in the summer months, but lower in other periods, especially in autumn and winter, this formula can reasonably adjust the total contribution share of these monthly figures downward or upward by considering the specific K_m of the month to achieve a balance.

[0086] The actual curtailment distribution curve is then predicted based on the adjusted power generation. The purpose of this step is to understand that even with the ability to optimize power estimation, there may still be electricity that cannot be fully utilized due to external factors. Curtailment refers to the inability of excess photovoltaic power to be transmitted to the grid due to load restrictions or other technical problems at the grid connection point. In actual use, this situation occurs when the installation area of the campus photovoltaic system is large and the peak power supply far exceeds the power demand. For example, in a specific scenario, calculations show that there will be local curtailment of photovoltaic power during the peak noon period for more than 30 days each year. This can remind designers to improve the access plan or plan energy storage to store surplus power during specific periods of time.

[0087] The above steps constitute the key detailed methodology for the entire photovoltaic potential assessment, laying a scientific foundation for accurate evaluation and subsequent improvement.

[0088] Next, we describe the present invention's method of comparing potential power generation with actual load curves to determine the distribution of curtailment rates. This process involves several specific steps: first, using a dynamic time-shift matching method to adjust the curve's time deviation, then calculating the daily net power difference using a formula, then determining whether the rated energy storage threshold is exceeded, and finally designing a time-sharing priority charging and discharging strategy for areas with high curtailment rates.

[0089] In the first step, the dynamic time-shift matching method is used to adjust the time deviation Δt of the two curves of potential power generation and actual power load. The purpose of this step is to correct the errors that may be caused by different data source acquisition mechanisms, ensure that the two curves correspond accurately on the time axis, and thus improve the accuracy of subsequent calculations. Among them, Δt is the time difference between the photovoltaic power generation curve and the power load curve, which usually ranges from -1 hour to +1 hour. For example, in one embodiment, by matching the sampling points within a day one by one, it was found that the maximum time shift difference between the two was 30 minutes, and the curve overlap was greatly improved through interpolation correction.

[0090] The second step is to measure the daily net difference in electricity consumption using the formula \(D_t = |\sum_{i=1}^N P_i L_i|\). In this formula, the parameters \(P_i\) and \(L_i\) represent the instantaneous values of photovoltaic power generation and actual electricity load, respectively, at each time step i; and N represents the total number of time intervals (for example, a day divided by hours has 24 steps). A larger value of \(D_t\) indicates a more pronounced supply-demand imbalance. Optimally, \(D_t\approx 0\) is desired. This formula is designed to comprehensively assess the total deviation in energy supply and demand per unit time, laying the foundation for subsequent judgments.

[0091] The next step is to determine whether there is an overload risk period. If \(D_t\) exceeds the preset energy storage device capacity C_th for multiple consecutive days on a certain day or period, it is considered to have entered the excessive power difference risk stage, and precautionary measures should be deployed. In this scenario, the set threshold \(C_th\) is selected based on the battery storage specifications and the average results of long-term historical load data. Specifically, for example, assuming that the current campus power storage system can withstand a maximum charging limit of 500kWh / d, after multiple statistics, it is found that the net difference exceeds this limit frequently around noon in the summer, which can be identified as a typical risk range.

[0092] Finally, to mitigate the exacerbated curtailment of solar power caused by the aforementioned situation, a time-of-use management strategy is proposed for areas experiencing peak sunlight and low electricity consumption. Specifically, when identifying specific areas where excess power is generated by arrays on the rooftops of research buildings but no immediate means of dissipating it, an intelligent scheduling module can be deployed in advance to optimize scheduling logic for more efficient resource utilization and reduced energy waste. For example, a software system combined with weather forecast models can be used to pre-plan specific times when more solar power should be diverted to the grid rather than directly supplied to end consumers, achieving a balanced energy balance.

[0093] Next, we will further describe the present invention's approach to constructing a campus-based energy storage system control model based on the distribution of curtailed solar power. This model specifically includes the following steps: setting the maximum capacity of the energy storage battery and its current energy storage level; establishing a constraint function to determine the curtailed solar power level; determining whether optimization is needed to reduce curtailed solar power; and enabling auxiliary energy to address power shortages. These steps comprehensively optimize the utilization and economic benefits of the photovoltaic system.

[0094] The first step is to set the maximum capacity S_MAX of the energy storage battery and record its current energy storage level H. The maximum capacity represents the upper limit of the maximum amount of electric energy that the energy storage system can store, and the unit is usually kWh (kilowatt-hour). This parameter is determined based on the photovoltaic installed capacity and the school's electricity load characteristics. It is generally selected to be appropriately expanded by about 1.2 times based on the peak-to-valley fluctuation range of photovoltaic power generation on campus throughout the year in order to cope with special circumstances. For example, in one embodiment, assuming that the rated total power of the photovoltaic power generation system installed on the campus is 50kW and the average daily power generation throughout the year is 200kWh, then S_MAX can be set to 240kWh, and the current energy storage level H represents the current power state, which is in the range of [0, S_MAX].

[0095] The second step is to calculate the real-time curtailed solar power using the formula J = max(0, G_t - L_t \cdot S_RATE \cdot R_H). G_t represents the instantaneous output power of the photovoltaic power generation at time t, in kW; L_t represents the load demand power at time t, also in kW; S_RATE is the charge and discharge rate limit of the energy storage device, the upper limit of the energy that can be processed per unit time, and the typical value range is [0, 1]; R_H is the multiplier factor related to the charging efficiency of the energy storage device, usually set in the range of [0.8, 0.9]. The significance of the formula design is that when the photovoltaic power supply exceeds the available storage capacity and the actual load requirements, this excess power is marked as potential curtailed solar power.

[0096] The third step is to evaluate whether optimization is needed to reduce the waste of curtailed solar power. The condition is that L_t+J>H+DIS_TH, where DIS_TH is defined as the lower limit of the allowable depth of discharge (i.e., the minimum amount retained to avoid depleting the battery), and the empirical value is usually set at 20% to 30%. When the sum of the photovoltaic energy generated and the predicted remaining curtailed solar power exceeds the sum of the existing energy storage plus the retention threshold, the optimization algorithm is activated to reduce losses, such as adjusting the equipment load curve or dispatching other backup plans in advance to ensure the maximum utilization of clean energy output while maintaining the balance and stability of the grid.

[0097] Finally, during low-sunlight conditions, such as at night or during periods of continuous rain, auxiliary energy supply mechanisms should be activated to ensure stable operation. Specifically, if a campus area frequently experiences insufficient sunshine for several consecutive days during winter, natural gas generators or remote power dispatch can be used as an emergency backup solution. Based on a cost-benefit assessment, the optimal time of use can be allocated to ensure the most economical and efficient system operation.

[0098] This demonstrates that by setting initial parameters, monitoring operational constraints, and developing emergency response measures, refined management and improved dynamic response capabilities have been achieved for distributed photovoltaic systems embedded in campus buildings. These measures will help further advance sustainable development goals and reduce energy dependence.

[0099] Next, the present invention is described in detail, which further formulates an optimized energy storage scheduling strategy based on the control model to reduce the abandonment rate of the photovoltaic system. According to the calendar periodic characteristics, a multi-stage energy storage strategy S_n is planned, and the energy storage coefficient k_n of each stage is defined. This step aims to combine the fluctuations in electricity demand in different seasons and time periods to set a reasonable charge and discharge ratio for the energy storage system to avoid overcharging or abandonment waste. For example, at the beginning and end of the spring semester, the campus electricity demand is low due to students' holidays or reduced activities; while the peak electricity consumption increases significantly after the start of the autumn semester. By reasonably allocating the energy storage coefficient, for example, k_n∈[0.1,1.0], the optimal value may be set in the range of about k_n≈0.7, which can find a balance between meeting the energy storage target and the load.

[0100] Based on the condition \(E_{store}(n+1)=E_{current}\cdot e^{k_d\cdot d}\), the decrease in remaining energy as the waiting time increases is recorded. This formula quantifies the decreasing trend of energy due to natural loss over a certain period of time. The parameter \(k_d\) represents the natural energy loss rate. In scenarios such as lithium batteries, the empirical value of \(k_d\) is typically between [0.001 and 0.003] per day, depending on the battery material. The parameter \(d\) represents the time span for energy storage. The formula is designed to leverage the exponential law of physical energy decay to accurately estimate the available energy within different retention periods.

[0101] A tolerance factor for curtailment of solar power, \(φ_h\), is set up separately for critical days (such as holidays), and the optimized control rule is obtained by combining it with the real-time weather index \(W_r\). The goal here is to dynamically adjust the system's acceptable curtailment of solar power during special periods (such as statutory holidays or low periods of campus activity). The tolerance factor \(φ_h\) defines a relatively loose threshold to measure the allowable level of wasted energy, and the common range of \(φ_h\) is [0.15, 0.30]. In addition, the real-time weather index \(W_r\) is introduced as an important factor affecting photovoltaic efficiency (\(W_r=1\) indicates an unobstructed sunny day, and \(W_r<1\) indicates shadows or cloud effects). The combined optimization results of the two can make the energy storage solution more suitable for unpredictable weather and holiday load reductions.

[0102] Under extreme irradiance fluctuations, temporary excess storage capacity is permitted to ensure a balanced and stable power supply. This means the energy storage system needs to retain buffer capacity to cope with sudden, dramatic changes in sunlight or other disruptions. For example, when a strong storm causes a sudden drop in sunlight, an additional 5%-10% of energy storage reserves can be used to compensate and adjust to ensure consistent output power.

[0103] In one example, an evaluation and strategy optimization were conducted for photovoltaic power generation on the rooftop of a university library. During the two weeks following the start of the spring semester, while sunlight was excellent, the load demand accounted for only approximately 60% of the power generation capacity, as students were not yet fully engaged in their daily activities. This resulted in significant curtailment of solar power. By setting an appropriate energy storage factor k_n≈0.8 and calculating the short-term storage effect based on the daily residual charge decay pattern, the control parameters were optimized based on real-time meteorological data and the May Day holiday period (φ_h≈0.28), ultimately achieving a desired reduction in curtailment of solar power by 23.7%.

[0104] Next, the present invention is described as follows:

[0105] The first step is to construct the objective function G(x,y) = R_h * F_g^y / x, taking into account historical PV utilization data (R_h) and the projected growth factor F_g for the next N years. In this step, x is a subjective optimization weight, ranging from [0.5 to 2]; the optimal value depends on the actual dataset distribution; y represents the incremental order, typically set to an integer between 1 and 5 to balance algorithm complexity and accuracy. The core meaning of this formula is to mathematically characterize the long-term benefit potential of the PV system based on existing historical data (R_h) and estimated future potential (F_g). This weighting is then used to adjust priorities, thereby achieving a scientific and rational resource allocation assessment.

[0106] The formula B_t = |G(t)P_d|·β_p is used to calculate the single-period energy storage dispatch cost B_t. P_d is the power demand in the current time slot, and the β_p conversion factor is determined by the differences in the time-of-use electricity price structure of the power grid. Specifically, for example, in a campus scenario, if the photovoltaic system power generation in a specific time period is not enough to fully match the load demand, it is necessary to introduce external electricity supplement, which in turn generates energy storage deployment expenses; β_p can be set according to the local peak and valley electricity price fluctuation characteristics, and its range is approximately 0.6 to 1.5 yuan / kWh. This definition ensures that economic feasibility is taken into consideration and avoids high energy consumption and low efficiency.

[0107] The curtailment penalty term P_l = η*(D_max / D_target-1)^2 is introduced into the overall scheduling algorithm to improve optimization. Here, D_max represents the theoretical upper limit of the maximum PV power generation capacity, and D_target represents the target value to be achieved under ideal conditions. The coefficient η is used to emphasize the cost of non-compliance, and the recommended initial setting range is [100, 300]. This measure ensures that excessive resource waste can be promptly corrected in actual operation, helping to improve overall efficiency.

[0108] Dynamic threshold update rules are required to maintain scheduling flexibility for special meteorological conditions. In one embodiment, treatment thresholds can be adaptively adjusted to address the risk of sudden cloud cover causing rapid reduction in sunlight or power outages caused by summer rainstorms and lightning strikes. For example, system response samples under similar conditions over the past five years can be collected in advance, and a statistical model can be established to predict the optimal critical parameter values corresponding to new input variables, thereby ensuring that relatively stable and efficient operation can be maintained even in extreme environments.

[0109] Next, the description of the present invention also includes the specific operating steps as follows:.

[0110] The first step involves extracting the photovoltaic load ratio of key nodes, \(M_p\), and the state vector of the energy storage station, \(X_s\), to locate the bottleneck. This step aims to screen out nodes that have a greater impact on overall efficiency by analyzing the data of each node of the campus building photovoltaic system, and determine whether there is an imbalance in their load conditions and the current use status of the energy storage device. Among them, \(M_p\) represents the ratio of photovoltaic power supply to demand load, generally between 0 and 1, and close to 1 indicates that the node is completely self-sufficient. The \(X_s\) vector contains key parameters such as the power status of each energy storage unit and its charging and discharging efficiency information.

[0111] In the second step, an iterative correction method is used to optimize the configuration formula \(ΔX_s(n) = \alpha*[\partial\Lambda(X_s) / \partial X_s]\) to fine-tune the state parameters of the storage station to achieve the desired energy efficiency target. The step size adjustment term α is typically set in the range of [0.01, 0.5] to prevent algorithm divergence, and an empirical constant such as 0.1 is recommended as the initial value. Λ, serving as the energy efficiency evaluation objective equation, reflects the comprehensive energy utilization index of the campus power system. The formula is designed to dynamically adjust the values of each element in the \(X_s\) parameter set to achieve the optimal value of the objective function while meeting hardware performance constraints.

[0112] In the subsequent search for multiple possible solution sets, a specific solution must be found that satisfies all constraints while minimizing curtailed power (i.e., excess photovoltaic power generation that exceeds the system's capacity and is not utilized promptly). In one embodiment, for example, several different time-slicing allocation strategies can be compared. Each strategy is simulated based on a typical daily campus load curve, and the total power consumption, remaining storage, and losses of each strategy are recorded. The optimal result is then selected for reference in the next deployment plan.

[0113] Finally, simulated feedback testing is performed to improve energy storage efficiency, ensuring that there is no risk of additional operating expenses due to redundancy or failure. For example, when certain energy storage configuration recommendations derived from preliminary calculations are actually applied, deviations may occur and require further correction. For example, independent analyses are conducted for two scenarios: summer with abundant sunlight and winter with limited sunlight resources. The stability of the final solution is then verified by combining common rules for both seasons. Specifically, new data from each experiment is input into the model and the aforementioned process is repeated to form a closed-loop learning process until a satisfactory output configuration is achieved.

[0114] Next, the present invention is described in particular. First, a thermal balance evaluation sub-process is proposed to optimize the heat dissipation effect under high temperature conditions during the day in summer. The core purpose of this step is to design a specific heat dissipation strategy and analysis process to address the problem that the higher ambient temperature on campus during the day in summer may cause the temperature of photovoltaic panels to rise. Specifically, through simulation and actual measurement methods, a quantitative evaluation of the heat dissipation performance of the photovoltaic panel surface is carried out, and corresponding adjustment measures are introduced to reduce the negative impact of heat accumulation.

[0115] Then, the ambient temperature and humidity coupling effect estimation module E_a=λ*exp((THTH_opt) / (δ_T))*S_max is introduced, where the parameter THTH_opt represents the set optimal reference temperature benchmark, TH represents the actually measured surface temperature of the photovoltaic panel, S_max is the theoretical value of the maximum power output point of the photovoltaic panel, δ_T and λ are key numerical parameters that characterize temperature sensitivity and equipment heat dissipation efficiency, respectively. λ is generally between 0.75 and 0.98, and the optimal value is adjusted depending on the specific equipment material performance. The optimal range is approximately 0.83 to 0.87; the typical range of δ_T is set between 1 and 15 degrees Celsius. The significance of this formula is to comprehensively evaluate how environmental factors affect actual output performance by combining temperature, heat dissipation efficiency and humidity, and ultimately estimate the potential reduction in energy output through the model.

[0116] In one embodiment, when the measured actual surface temperature TH_ex exceeds a predetermined limit TH_lim (typically 65°C, the recommended upper limit for stable PV system operation) or the current relative humidity exceeds a safety threshold, cooling loss reduction intervention is immediately executed to prevent output efficiency degradation and the resulting energy waste caused by continued rising module temperatures. This operation ensures that even in extreme environments, efficient and stable operation can be maintained to protect initial profit targets.

[0117] Specifically, the rational design of ventilation layouts further enhances the role of natural wind flow organization to reduce the likelihood of localized dead zones. For example, for rooftop photovoltaic arrays, the addition of guide fences alters the angles of the existing turbulent airflow, ensuring more uniform coverage across the entire working surface and improving heat dissipation efficiency. These measures ensure that the overall system functions smoothly and maintains expected performance even during significant external climate fluctuations, achieving the ideal goal of long-term stable operation.

[0118] Next, the additional steps of the present invention are described. First, long-term trend fitting ability in a complex environment is mined through deep learning to form a decision basis matrix A_pred. By constructing a deep learning model, long-term trend prediction is carried out on the possible operating conditions and energy output conditions of the photovoltaic system in a complex environment. Specifically, through model analysis of the dynamic evolution of variables such as light intensity, temperature, and cloud distribution in the time dimension, the predicted values of the photovoltaic power generation efficiency under each parameter combination are obtained, and the matrix A_pred is generated to represent the potential photovoltaic power generation results under various input conditions. For example, in a method for evaluating the photovoltaic potential of a campus building complex, using meteorological data and actual power generation data of the past year to train the model can accurately evaluate the quantifiable values of the electric power contributed by roofs with different orientations and slopes in each campus under rainy seasons, cloudy days, and other complex climate conditions.

[0119] Subsequently, the photovoltaic load imbalance risk level scoring index R_s = [∑|G_iL_j|i<j / MAX(|GiGavg||)]*ξ is defined, where G_i is the photovoltaic power generation amount value of each region at a specific moment, Gavg is the historical power generation average value in the region, the MAX function is used to represent the maximum difference amplitude range between regions, |G_iL_j| is the absolute value of the difference between the power supply amount of a single region and its required power load, and ξ is the normalized scaling factor. The purpose of this formula setting is to measure the overall power balance degree and its potential risks by quantifying the deviation between the energy supply and demand of each section within each time period. In one embodiment, when the power generation of a certain building on campus drops rapidly due to cloud cover after the peak sunlight in the afternoon and cannot meet the immediate electricity demand, a high imbalance score value will be generated by this formula calculation to identify such time periods that need attention.

[0120] After that, the region with the highest probability is selected through the activation layer as the action guidance for the next day and the preset action list L_optimal is generated. Specifically, this is a process of using a specific structure (such as softmax activation) in the neural network to identify and select the highest possible option in the future state prediction. These possibilities will be translated into operation suggestions according to the above two-step calculations, covering key task arrangements such as adjusting the power output distribution plan or starting and stopping energy storage devices. For example, when it is predicted that the temperature will be low and there will be mist in the morning of the next day, affecting the power generation efficiency, the battery filling capacity should be increased or other energy supplement mechanisms should be mobilized in advance to enter the operation mode, and recorded in the preset list for real-time retrieval and application.

[0121] Furthermore, when sudden, drastic changes in light intensity occur, the system promptly switches to the most adaptable emergency response mode to minimize the magnitude of temporary disturbance losses. This means a rapid feedback mechanism has been designed to instantly activate the optimal emergency plan in unusual circumstances. If, for example, a sudden storm cloud blocks light, causing a sudden drop in power generation, the previously defined response plan is automatically implemented to immediately adjust the distribution system configuration to mitigate the impact of the sudden change. This ensures stable operation of the entire system while maintaining efficient operating levels without disruption.

[0122] Next, the description of the present invention also involves supplementary details: First, an anomaly detection module is added to the evaluation method. This link refers to the introduction of an intelligent mechanism to monitor the system operation status and generate a response instruction set \(M_{\text{repair}}\) for abnormal equipment operation to guide operation and maintenance operations. Secondly, a probability-based fault location method is adopted to calculate the influence ratio formula of each possible problem part \(P_i=f(i)\times\exp(d_i / c)\), which involves key parameters and their definition domains, and the final conclusion is drawn to provide a basis for the maintenance list. Finally, the overall system is continuously monitored and maintained and optimized to ensure efficient energy utilization and prevent waste.

[0123] Specifically, the addition of an anomaly detection module enables automated alarm response by collecting and analyzing data from core components such as sensors and inverters in campus building photovoltaic systems. It uses preset rules or machine learning models to filter out abnormal indicator values to trigger alarms and automatically generates a response instruction set, M_{\text{repair}}\. For example, if monitoring detects a sudden drop in the output power of a photovoltaic array, the module can identify this event as a decrease in efficiency due to dust accumulation and add a cleaning prompt to the response set. This step is intended to improve fault response speed and reduce human judgment bias.

[0124] Regarding the core formula of the fault location probability derivation method, \(P_i=f(i)\times\exp(d_i / c)\), the deceleration factor \(f(i)\) reflects the proportional characteristic of the degradation rate of the i-th component decreasing over time (the value is usually in the range of 0 to 1), the exponential factor part variable d represents the actual deviation of the component damage from the expected performance value, and c represents the critical standard threshold limit parameter. It is recommended to set c between the numerical range of 3-5 to obtain a balance between the accuracy of the result and the sensitivity level. The formula logic design is based on the assumption that the more serious the deviation, the more likely there is a potential hidden danger, and takes into account the impact of the accumulated wear history of the equipment. In one embodiment, if the temperature of the connection line of a distribution cabinet exceeds the rated range by a large amount (large di value), its high-risk possibility ranking in the list of all suspected fault source locations can be determined according to the calculation.

[0125] The key output from integrating all diagnostic information into a comprehensive repair checklist is the two aforementioned steps. This document not only identifies the precise problem areas but also provides an estimate of the resources and time required for repair, making it easy for engineering staff to reference. For example, a final checklist presented to the maintenance team prioritizes clearing the south-facing photovoltaic panels in the main control building. This easy-to-understand action guide significantly accelerates the repair process and reduces waiting times.

[0126] Comprehensive monitoring measures ensure that every component of the photovoltaic system maintains optimal performance, including adjusting power generation parameters to match load usage and avoid unnecessary losses. Specifically, this is achieved through intelligent scheduling, such as adjusting the charging and discharging strategies of energy storage units or switching inefficient components off the main line.

[0127] The parametric analysis-based method for assessing the photovoltaic potential of campus buildings includes a comprehensive analysis of the photovoltaic systems on campus rooftops, encompassing a series of technical steps, from assessing photovoltaic power generation potential to optimizing energy storage scheduling strategies, thereby achieving scientific management and reduction of the photovoltaic system's curtailment rate. First, the potential power generation of the photovoltaic systems on campus rooftops is analyzed by analyzing the year-round variations in irradiation intensity. This process involves collecting and analyzing years of historical climate data for the region, identifying the intensity and variation of solar irradiation at different times of the day, and establishing a correlation model between the photovoltaic system and weather conditions, thereby accurately estimating the potential power generation for the entire year.

[0128] Next, the predicted potential PV power generation is compared with the campus' actual electricity load curve to assess whether PV output matches actual electricity demand and determine the distribution of curtailment rates within each time period. This phase reveals the extent of energy waste caused by insufficient load during peak PV generation periods, providing an important basis for subsequent strategy formulation. Based on this calculated curtailment rate distribution, a storage system control model suitable for campus scenarios is constructed, using mathematical modeling to simulate the impact of factors such as different storage capacities and storage charge and discharge powers.

[0129] Finally, based on the results of the above model, a specific optimized energy storage scheduling strategy was developed for the photovoltaic system's operation. This strategy dynamically adjusts the operating mode of the energy storage equipment, such as increasing the energy storage charging efficiency during periods of high photovoltaic power generation or adjusting the discharge behavior during periods of high and low power, thereby significantly reducing the phenomenon of curtailment caused by excessive power generation. In short, this method solves the problem of how to accurately design energy storage control measures based on the changing characteristics of irradiation intensity throughout the year to reduce energy waste in campus building photovoltaic systems.

[0130] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present invention. These improvements and modifications should also be regarded as the scope of protection of this application.

Claims

1. A method for evaluating photovoltaic potential of campus buildings based on parametric analysis, characterized by: include: The potential power generation of the photovoltaic system on the roof of campus buildings is obtained based on the analysis of the annual radiation intensity variation pattern; Compare and analyze the potential power generation and actual power load curve to determine the distribution of curtailment rate; Based on the distribution of the abandoned solar power rate, a storage system control model suitable for campus scenarios is constructed; An optimized energy storage scheduling strategy is formulated based on the regulation model to reduce the curtailment rate of the photovoltaic system.

2. The campus building photovoltaic potential assessment method based on parametric analysis according to claim 1 is characterized in that: The potential power generation of the campus building rooftop photovoltaic system based on the analysis of the annual radiation intensity variation pattern further includes: Model the daily average radiation intensity G based on different seasons; The formula for calculating the total annual radiation energy Q is Q = \int_{0}^{365}G(t)\cdot dt, where t is the cumulative number of days in a year and G(t) is the average daily radiation intensity on day t; A threshold T is set during the potential light abandonment period. When G(t)>T, the period is marked as a high irradiance output period. A local control energy storage solution is constructed for the high irradiation output section.

3. The campus building photovoltaic potential assessment method based on parametric analysis according to claim 2 is characterized in that: The following steps are also included: Correct the potential power generation by analyzing the peak irradiation intensity P_m month by month throughout the year; Calculate the monthly average peak adjustment ratio K_m = (P_max P_m) / P_max, where P_max represents the maximum peak radiation intensity of the whole year; The formula C = \sum_{m=1}^{12}G_m\cdot K_m is introduced to calibrate the compensation of the total potential power generation for the whole year, where G_m represents the average potential power generation in the mth month; The actual curtailment distribution curve is predicted based on the adjusted power generation.

4. The method for evaluating campus building photovoltaic potential based on parametric analysis according to claim 3 is characterized in that: The comparative analysis of the potential power generation and the actual power load curve to determine the distribution of the abandoned solar power rate further includes: The dynamic time-shift matching method is used to adjust the time deviation Δt of the two curves so that they can coincide more accurately. Use the formula D_t = |\sum_{i=1}^N P_i L_i|, where P_i and L_i represent the values of photovoltaic power generation and actual power load at time step i respectively; If the daily net difference in energy consumption D_t continuously exceeds the rated energy storage threshold C_th, this time period is marked as an overload risk period; Design a time-sharing priority charging and discharging strategy for areas with high curtailment rates to reduce losses.

5. The method for evaluating campus building photovoltaic potential based on parametric analysis according to claim 4 is characterized in that: The step of constructing an energy storage system control model suitable for campus scenarios based on the abandoned solar power rate distribution further includes: Set the maximum capacity S_MAX of the initial energy storage battery and record the current energy storage level H; Establish a constraint function J = max(0, G_t L_t S_RATE\cdot R_H) to determine the real-time abandoned optical power, where S_RATE is the storage rate and R_H is the current charging rate. If L_t+J>H+DIS_TH, the optimization mechanism will be triggered to reduce abandoned light; Enable auxiliary energy supplement function during low sunlight period to alleviate power shortage problem.

6. The method for evaluating campus building photovoltaic potential based on parametric analysis according to claim 5, characterized in that: Formulating an optimized energy storage scheduling strategy based on the control model to reduce the abandonment rate of the photovoltaic system further includes: Plan a multi-stage energy storage strategy S_n based on the calendar periodicity characteristics and define the energy storage coefficient k_n for each stage; Based on the condition \(E_{store}(n + 1)=E_{current}\cdot e^{k_d\cdot d}\), record the reduction of the remaining electric energy with the waiting time. Here, \(k_d\) represents the natural loss rate, and \(d\) represents the number of days of time lag. Separate the light curtailment tolerance factor \(\varphi_h\) for key days (such as holidays), and combine it with the real-time weather index \(W_r\) to obtain the optimized control rule. Allow the energy storage to be overused for a short time under extreme irradiance fluctuations to ensure balanced and stable power supply.

7. The method for evaluating campus building photovoltaic potential based on parametric analysis according to claim 6, characterized in that: It also includes the following steps: Comprehensively consider the historical photovoltaic utilization rate \(R_h\) data and the future \(N\)-year prediction growth factor \(F_g\) to construct the objective function \(G(x,y)=R_h*F_g^y / x\), where \(x\) is the subjective optimization weight and \(y\) represents the increasing order. Calculate the single-period energy storage scheduling cost \(B_t\) using the formula \(B_t = |G(t)P_d|\cdot\beta_p\), and the conversion coefficient \(\beta_p\) depends on the factors of the electricity fee structure difference. Introduce the light curtailment penalty term \(P_l=\eta*(D_{max} / D_{target1})^2\) into the overall scheduling algorithm to improve the optimization effect. Set the dynamic threshold update rule for special meteorological conditions to maintain the scheduling flexibility.

8. The method for evaluating campus building photovoltaic potential based on parametric analysis according to claim 7, characterized in that: It also includes the following specific operation steps: Extract the photovoltaic load ratio \(M_p\) of the key nodes and the state vector \(X_s\) of the energy storage station to locate the bottleneck link. Use the iterative correction method to optimize the configuration formula \(\Delta X_s(n)=\alpha*[\partial\Lambda(X_s) / \partial X_s]\), where \(\alpha\) is the step adjustment term and \(\Lambda\) is the energy efficiency evaluation target equation. Find the optimal ratio value that satisfies the constraint conditions and minimizes the discarded electricity among multiple sets of possible solutions. Implement the simulation feedback test to improve the energy storage allocation efficiency and prevent the risks of redundancy or failure.

9. The method for evaluating campus building photovoltaic potential based on parametric analysis according to claim 8, characterized in that: In particular, Propose a thermal balance evaluation sub-process for optimizing the heat dissipation effect under the high-temperature conditions during the day in summer. Introduce the environmental temperature and humidity coupling effect estimation module \(E_a=\lambda*exp((THTH_{opt}) / (\delta_T))*S_{max}\), where \(TH\) is the temperature reference benchmark and \(\lambda\) is the heat dissipation efficiency parameter. When the measured actual surface temperature \(TH_{ex}>TH_{lim}\) or the relative humidity exceeds the limit, immediately execute the refrigeration and loss reduction measures to reduce the energy waste phenomenon caused by the performance degradation of the photovoltaic panel. Improve the air circulation efficiency by reasonably designing the air duct direction, so as to ensure that the overall operation stability is not overly disturbed by the climate conditions.

10. The campus building photovoltaic potential assessment method based on parametric analysis according to claim 9 is characterized in that: An additional step is added: Use deep learning to挖掘 the long-term trend fitting ability in complex environments to form the decision basis matrix \(A_{pred}\). Define the photovoltaic load imbalance risk level scoring index \(R_s=[\sum|G_iL_j|i < j / MAX(|GiG_{avg}||)|]*\xi\), where \(\xi\) is the normalized scaling ratio factor. Select the area with the highest probability through the activation layer as the action guidance for the next day and generate the preset action list \(L_{optimal}\). When a sudden and剧烈 change in light occurs, promptly switch to the most adaptable emergency response mode to minimize the amplitude of the temporary disturbance loss.

Images