Energy risk assessment method for comprehensive energy system of agricultural park

Abstract

The invention relates to the technical field of park energy management, in particular to an energy risk assessment method for an agricultural park integrated energy system, which comprises the following steps: S1, data collection and preprocessing: collecting multi-dimensional data in an agricultural park; s2, microclimate-driven energy demand prediction: analyzing the dynamic influence of microclimate change on the energy demand of irrigation, refrigeration and light supplement, and forming energy demand prediction under different climate conditions; s3, energy consumption feedback of land degradation: analyzing extra energy demands caused by land degradation, and providing a relationship between a land degradation degree and energy consumption; and S4, comprehensive energy risk assessment: performing comprehensive assessment on the energy risk, and generating a risk early warning signal. According to the invention, the anti-risk capability and the operation efficiency of the park are improved, and a scientific basis and a management tool are provided for sustainable development of the agricultural park.

Description

Technical Field

[0001] The present invention relates to the technical field of park energy management, and in particular to an energy risk assessment method for an integrated energy system in an agricultural park. Background Art

[0002] With the intensification of global climate change and resource shortage problems, the sustainable development of agricultural parks has received more and more attention. The energy demand in agricultural parks involves irrigation, refrigeration, supplementary lighting and other aspects, which are not only closely related to microclimate changes, but also affected by the degree of land degradation. At the same time, due to the uncertainty of climatic conditions, the energy demand of the park is highly volatile. The rational allocation and efficient utilization of energy resources have become the core goals of park management. Therefore, how to achieve scientific management and risk assessment of the comprehensive energy system of agricultural parks under changing environmental conditions has become an urgent problem to be solved in the current agricultural field.

[0003] In the energy demand forecasting and risk assessment of agricultural parks, existing technologies usually only consider a single climate factor or simple energy consumption data, which makes it difficult to effectively capture the dynamic impact of microclimate and land degradation on energy demand. In addition, existing energy demand forecasting models often lack systematic processing and refined analysis of data, resulting in poor accuracy and real-time performance of forecast results. At the same time, existing energy risk assessments mostly rely on experience-based judgments and lack systematic risk quantification and early warning mechanisms, which makes it difficult for managers to respond to sudden changes in energy demand in a timely manner, increasing resource waste and operational risks.

[0004] The purpose of the present invention is to provide an energy risk assessment method for the integrated energy system of an agricultural park, which can not only improve the accuracy and stability of the park's energy management, but also support the park's rational allocation of resources under different environmental conditions, thereby enhancing risk resistance and sustainable development levels. Summary of the invention

[0005] The invention provides an energy risk assessment method for an agricultural park comprehensive energy system.

[0006] An energy risk assessment method for an integrated energy system in an agricultural park comprises the following steps:

[0007] S1, data collection and preprocessing: collect multidimensional data required for energy risk assessment of the integrated energy system in the agricultural park, including energy consumption data, microclimate data (temperature, humidity, wind speed, light) and land degradation data (soil salinization, erosion degree), and preprocess the multidimensional data, including data cleaning and standardization;

[0008] S2, Microclimate-driven energy demand forecasting: Based on the collected microclimate data, the dynamic impact of microclimate changes on the energy demand for irrigation, cooling and supplementary lighting is analyzed to form energy demand forecasts under different climate conditions;

[0009] S3, Energy consumption feedback of land degradation: Based on the collected land degradation data, analyze the additional energy demand caused by land degradation and provide the relationship between the degree of land degradation and energy consumption, including:

[0010] S31, land degradation feature extraction: based on the collected land degradation data, extract land degradation features, including soil salinization index and erosion grade;

[0011] S32, Modeling of additional energy demand for degraded land: Based on the extracted land degradation characteristics, an additional energy demand model for degraded land is constructed;

[0012] S33, Relationship mapping between land degradation and energy consumption: Through the additional energy demand model of degraded land, the relationship mapping between land degradation degree and energy consumption is analyzed, and the impact curve of different land degradation levels on energy demand is drawn to provide a relationship diagram between land degradation and energy consumption;

[0013] S4, comprehensive energy risk assessment: Based on the pre-processed energy consumption data, combined with the results of microclimate-driven energy demand forecast and energy consumption feedback from land degradation, a comprehensive assessment of energy risks is conducted and risk warning signals are generated.

[0014] Optionally, the data collection and preprocessing in S1 includes:

[0015] S11, Energy consumption data collection: Through smart electricity meters, smart gas meters, and flow meters, various energy consumption data in the agricultural park are collected in real time, including electricity, gas, and water consumption data, forming hourly or daily energy consumption records;

[0016] S12, microclimate data collection: deploy temperature, humidity, wind speed, and light environment sensors in the agricultural park to collect microclimate data, including temperature, humidity, wind speed, and light;

[0017] S13, land degradation data collection: soil sensors and remote sensing imaging technology are used to obtain soil salinization and erosion data, and combined with the geographic information system (GIS) to divide the land into plots and collect land degradation data in different areas;

[0018] S14, data cleaning: Use the outlier detection algorithm (3σ criterion) to identify and remove abnormal data points, normalize the multidimensional data, and scale all data values ​​to the [0,1] interval.

[0019] Optionally, the microclimate-driven energy demand prediction in S2 includes:

[0020] S21, microclimate feature extraction: based on the collected microclimate data, extract key climate features related to energy demand, including cumulative temperature, humidity index, sunshine hours, and wind speed change rate;

[0021] S22, microclimate-driven energy demand model construction: Based on the extracted key climate features, a microclimate-driven energy demand model is constructed;

[0022] S23, demand output: The energy demand for irrigation, cooling and lighting under different microclimate conditions is analyzed through the microclimate-driven energy demand model.

[0023] Optionally, the microclimate feature extraction in S21 includes:

[0024] S211, cumulative temperature: by accumulating the daily average temperature, the cumulative temperature within a certain period of time is calculated to reflect the heat demand for crop growth;

[0025] S212, Humidity Index: Combine relative humidity and air temperature to calculate the humidity index to reflect the impact of crops on irrigation needs;

[0026] S213, sunshine hours: The sunshine hours are recorded by the weather station, reflecting the effectiveness of light, and recorded in hours;

[0027] S214, wind speed change rate: calculates the degree of change of wind speed within a specified time period, reflecting the impact of wind speed on crop evaporation and cooling demand.

[0028] Optionally, the microclimate-driven energy demand model in S22 adopts an autoregressive integrated moving average (ARIMA) model, and the autoregressive integrated moving average (ARIMA) model includes:

[0029] S221, Feature selection and modeling objectives: Select microclimate characteristics (cumulative temperature, humidity index, sunshine hours, and wind speed change rate) that affect energy demand as exogenous variables and construct an autoregressive integrated moving average (ARIMA) model;

[0030] S222, difference processing: performing difference processing on the time series data to eliminate non-stationarity;

[0031] S223, model parameter identification: using the autocorrelation function (ACF) and partial autocorrelation function (PACF) to determine the order of the model;

[0032] S224, Model fitting and validation: Use the least squares method to estimate parameters and perform model validation to ensure goodness of fit;

[0033] S225, prediction and result output: use the fitted model to predict future energy demand and output the results.

[0034] Optionally, the demand output in S23 includes:

[0035] S231, Demand Decomposition: Based on Forecasted Energy Demand Allocate according to different demand types (irrigation, cooling, supplementary lighting);

[0036] S232, result output: Different microclimate characteristics are input through the microclimate-driven energy demand model, and the predicted values ​​of energy demand for irrigation, cooling and supplementary lighting under different microclimate conditions are output.

[0037] Optionally, the land degradation feature extraction in S31 includes:

[0038] S311, calculation of soil salinization index: According to the concentration of soluble salt in the soil, the soil salinization index SI is calculated to reflect the impact of salinization on land productivity;

[0039] S312, erosion level calculation: Determine the erosion level EL based on the volume loss rate of soil erosion.

[0040] Optionally, the additional energy demand model for degraded land in S32 adopts a Bayesian regression model, and the Bayesian regression model includes:

[0041] S321, introducing microclimate data: Based on the land degradation characteristics, temperature and humidity are combined as additional variables to build a Bayesian regression model;

[0042] S322, introduction of prior distribution: different prior distributions are set according to the importance of land degradation characteristics and additional variables;

[0043] S323, establishing a hierarchical Bayesian model: considering the differences between different plots, the Bayesian model parameters of each plot are estimated through a hierarchical model;

[0044] S324, parameter update and prediction through posterior distribution: update the parameters using Bayesian theorem and calculate the posterior distribution;

[0045] S325, Future Energy Demand Forecasting: Using the expected value of the posterior distribution to predict future energy demand.

[0046] Optionally, the relationship mapping between land degradation and energy consumption in S33 includes:

[0047] S331, Relationship mapping analysis: Based on the Bayesian regression model, the energy demand at each degradation level is calculated;

[0048] S332, Relationship diagram output: Draw a curve of energy demand varying with the degree of land degradation based on the results of relationship mapping analysis, and provide the generated relationship diagram of land degradation and energy demand to the agricultural park manager.

[0049] Optionally, the comprehensive energy risk assessment in S4 includes:

[0050] S41, Risk index calculation: Calculate the comprehensive energy risk index ERI based on energy consumption data, microclimate conditions, and the degree of land degradation.

[0051] S42, Risk level determination: Compare the calculated risk index ERI with the preset upper risk threshold T2 and lower risk threshold T1. When ERI < T1, it indicates low risk; when T1 ≤ ERI < T2, it indicates medium risk; when ERI ≥ T2, it indicates high risk.

[0052] S43, Generate risk warning signals: When medium risk or high risk is reached, generate risk warning signals and send warning messages to the park manager.

[0053] Advantages of the present invention:

[0054] In the present invention, through systematic multi-dimensional data collection and preprocessing, the comprehensiveness and accuracy of the data required for energy risk assessment in the agricultural park are ensured. Through the real-time collection of energy consumption data, microclimate data, and land degradation data, supplemented by data cleaning and standardization processing, data from different sources can be integrated on the same platform, eliminating the dimension difference and ensuring the high quality and comparability of the data.

[0055] In the present invention, through the microclimate-driven energy demand prediction and the additional energy consumption feedback model for degraded land, a dynamic and refined analysis of energy demand is achieved. By extracting key climate characteristics, constructing ARIMA models and Bayesian regression models, and combining the specific impacts of microclimate and land degradation on energy demand, the changing trends of irrigation, cooling, and supplementary lighting demands can be accurately predicted. Through demand decomposition and quantitative analysis, the energy demand is further refined, providing precise guidance for resource allocation in the park, reducing energy waste, improving resource utilization efficiency, and promoting a win-win situation for the ecological and economic benefits of the park.

[0056] The present invention can effectively identify and warn of potential energy risks by dynamically calculating the energy risk index through comprehensive analysis of microclimate conditions, land degradation degree and actual energy consumption, set reasonable risk thresholds through historical data statistics, and generate risk warning signals, thereby providing managers with forward-looking decision-making support and ensuring the stability and sustainability of park energy management. Overall, it not only improves the park's risk resistance and operational efficiency, but also provides a scientific basis and management tools for the sustainable development of agricultural parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A schematic diagram of the evaluation method flow in an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of energy consumption feedback for land degradation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0061] like Figure 1-Figure 2 As shown, a method for energy risk assessment of an integrated energy system in an agricultural park comprises the following steps:

[0062] S1, data collection and preprocessing: collect multidimensional data required for energy risk assessment of the integrated energy system in the agricultural park, including energy consumption data, microclimate data (temperature, humidity, wind speed, light) and land degradation data (soil salinization, erosion degree), and preprocess the multidimensional data, including data cleaning and standardization;

[0063] S2, Microclimate-driven energy demand forecasting: Based on the collected microclimate data, the dynamic impact of microclimate changes on the energy demand for irrigation, cooling and supplementary lighting is analyzed to form energy demand forecasts under different climate conditions;

[0064] S3, Energy consumption feedback of land degradation: Based on the collected land degradation data, analyze the additional energy demand caused by land degradation and provide the relationship between the degree of land degradation and energy consumption, including:

[0065] S31, land degradation feature extraction: based on the collected land degradation data, extract land degradation features, including soil salinization index and erosion grade;

[0066] S32, Modeling of additional energy demand for degraded land: Based on the extracted land degradation characteristics, an additional energy demand model for degraded land is constructed;

[0067] S33, Relationship mapping between land degradation and energy consumption: Through the additional energy demand model of degraded land, the relationship mapping between land degradation degree and energy consumption is analyzed, and the impact curve of different land degradation levels on energy demand is drawn to provide a relationship diagram between land degradation and energy consumption;

[0068] S4, comprehensive energy risk assessment: Based on the pre-processed energy consumption data, combined with the results of microclimate-driven energy demand forecast and land degradation energy consumption feedback, a comprehensive assessment of energy risks is conducted and risk warning signals are generated;

[0069] Through the above content, key data in the agricultural park are systematically collected and processed, and combined with the impact of microclimate and land degradation on energy demand and consumption, an accurate energy demand forecasting and consumption feedback model is constructed, which ultimately realizes the dynamic assessment and early warning of comprehensive energy risks, ensures the accuracy and sustainability of energy management, and helps the park optimize energy distribution under different environmental conditions, improve risk resistance and resource utilization efficiency.

[0070] Data collection and preprocessing in S1 include:

[0071] S11, Energy consumption data collection: Through smart electricity meters, smart gas meters, and flow meters, various energy consumption data in the agricultural park are collected in real time, including electricity, gas, and water consumption data, forming hourly or daily energy consumption records;

[0072] S12, microclimate data collection: deploy temperature, humidity, wind speed, and light environment sensors in the agricultural park to collect microclimate data, including temperature, humidity, wind speed, and light;

[0073] S13, land degradation data collection: soil sensors and remote sensing imaging technology are used to obtain soil salinization and erosion data, and combined with the geographic information system (GIS) to divide the land into plots and collect land degradation data in different areas;

[0074] S14, data cleaning: use the outlier detection algorithm (3σ criterion) to identify and remove abnormal data points, expressed as:

[0075]

[0076] Where x is the data value, μ is the mean, and σ is the standard deviation;

[0077] S15, data standardization: In order to eliminate the dimensional differences between different data dimensions, the multidimensional data is normalized and all data values ​​are scaled to the [0,1] interval, expressed as:

[0078]

[0079] Among them, x norm is the standardized data value, x min and x max They are the minimum and maximum values ​​of the dimension data respectively;

[0080] Through the above content, we have achieved comprehensive acquisition of energy consumption, microclimate and land degradation data in agricultural parks, ensuring the accuracy and real-time nature of the data. Data cleaning and standardization processing have eliminated outliers and dimensional differences, allowing various types of data to be effectively compared and analyzed on the same platform.

[0081] Microclimate-driven energy demand forecasts in S2 include:

[0082] S21, microclimate feature extraction: based on the collected microclimate data, extract key climate features related to energy demand, including cumulative temperature, humidity index, sunshine hours, and wind speed change rate;

[0083] S22, microclimate-driven energy demand model construction: Based on the extracted key climate features, a microclimate-driven energy demand model is constructed;

[0084] S23, demand output: through the microclimate-driven energy demand model, the energy demand for irrigation, cooling and supplementary lighting under different microclimate conditions is analyzed;

[0085] Through the above content, the scientificity and accuracy of the model construction are ensured, which can effectively analyze the irrigation, cooling and supplementary lighting needs under different microclimate conditions, provide real-time and accurate energy demand forecasts for agricultural parks, and then optimize resource allocation and reduce energy waste. In addition, it can enhance the park's adaptability to climate change, improve the sustainability of agricultural production, and promote a win-win situation of ecological and economic benefits.

[0086] Microclimate feature extraction in S21 includes:

[0087] S211, cumulative temperature: by accumulating the daily average temperature, the cumulative temperature within a certain period of time is calculated to reflect the heat demand for crop growth, expressed as:

[0088]

[0089] Where CT is the cumulative temperature, T i is the average temperature on the i-th day, and n is the number of days in the time period;

[0090] S212, Humidity Index: Combine relative humidity and air temperature to calculate the humidity index to reflect the impact of crops on irrigation needs, expressed as:

[0091]

[0092] Among them, HI is humidity index, RH is relative humidity (percentage), and T is temperature (degrees Celsius);

[0093] S213, sunshine hours: The sunshine hours are recorded by the weather station, reflecting the effectiveness of light, and recorded in hours;

[0094] S214, wind speed change rate: calculates the degree of change of wind speed within a specified time period, reflecting the impact of wind speed on crop evaporation and cooling demand, expressed as:

[0095]

[0096] Where VCR is the wind speed change rate, V i is the wind speed at the i-th moment, is the average wind speed during the time period, and n is the number of data points;

[0097] Through the above content, it is possible to clearly reveal the impact of climate characteristics such as cumulative temperature, humidity index, sunshine hours and wind speed change rate on irrigation, cooling and supplementary lighting needs, ensuring a deep understanding of the dynamic changes in energy demand. Through scientific calculation formulas and data analysis, agricultural parks can achieve more efficient resource allocation and management, thereby optimizing energy use, reducing waste, increasing crop yield and quality, and promoting sustainable agricultural development.

[0098] The microclimate-driven energy demand model in S22 adopts the autoregressive integrated moving average (ARIMA) model, which includes:

[0099] S221, Feature selection and modeling objectives: Select microclimate characteristics that affect energy demand (accumulated temperature, humidity index, sunshine hours and wind speed change rate) as exogenous variables and construct an autoregressive integrated moving average (ARIMA) model, which is expressed as:

[0100]

[0101] Among them, E t is the energy demand at time t, E t-1 、E t-2 are the actual energy demand at time t-1 and time t-2, is the autoregressive coefficient, which indicates the influence of past energy demand on current demand. θ1 and θ2 are sliding average coefficients, which indicate the influence of past error terms on current demand. t is the white noise error term, which indicates the random fluctuations that the model fails to capture, ∈ t-1 ,∈ t-2 Represent the white noise error term of the previous moment and the previous two moments respectively, X t-i is an exogenous variable (microclimate characteristics), representing the climate factors related to energy demand, β j is the regression coefficient of the exogenous variable, indicating the influence of each exogenous variable on energy demand, and n is the number of exogenous variables;

[0102] S222, difference processing: perform difference processing on time series data to eliminate non-stationarity, expressed as:

[0103] D t =E t -E t-1 ;

[0104] Among them, D t For the differenced data, ensure the model adapts to trend and seasonal changes;

[0105] S223, model parameter identification: The order of the model is determined using the autocorrelation function (ACF) and the partial autocorrelation function (PACF), expressed as:

[0106]

[0107] Among them, ACF(k) is the autocorrelation coefficient of time lag k, which indicates the correlation between energy demand and its lagged value. PACF(k) is the partial autocorrelation coefficient of time lag k, which indicates the correlation between the current value and the lagged value when controlling the first k-1 lagged values. is the mean of energy demand, N is the sample size, which represents the total number of time series data points;

[0108] S224, model fitting and verification: Use the least squares method to estimate parameters and perform model verification to ensure goodness of fit, expressed as:

[0109]

[0110] in, is the energy demand forecast value at time t after fitting, is the estimated model parameter, which represents the impact of each factor in the model on energy demand;

[0111] S225, prediction and result output: Use the fitted model to predict future energy demand and output the results, expressed as:

[0112]

[0113] in, is the predicted value of energy demand at time t+h, h is the predicted time step, indicating the predicted distance from the current moment to the future, are the predicted energy demand values ​​at the future prediction time points t+h-1 and t+h-2, respectively, t+h-i is the value of the jth exogenous variable at time t+hi;

[0114] Through the above content, exogenous variables are introduced to more accurately capture the dynamic impact of microclimate characteristics on energy demand. Not only the autocorrelation of time series is taken into account, but also the real-time impact of climate factors is effectively integrated, making energy demand forecasts more accurate and reliable. It can help agricultural parks optimize energy allocation under different microclimate conditions, improve resource utilization efficiency, and reduce operating costs, thereby supporting sustainable energy management and production.

[0115] The demand output in S23 includes:

[0116] S231, Demand Decomposition: Based on Forecasted Energy Demand According to different demand types (irrigation, cooling, supplementary lighting), it is allocated as follows:

[0117]

[0118] in, are the energy requirements for irrigation, cooling and supplementary lighting, respectively, irrigation , α cooling , α lig h ting are the weights of irrigation, cooling and supplementary lighting requirements, α irrigation +α cooling +α lig h ting =1;

[0119] The weights for irrigation, cooling, and lighting are set based on historical energy demand data for different microclimates, including:

[0120] Statistical analysis based on historical data: By analyzing past energy consumption data, determine the proportion of each demand type (irrigation, cooling, supplementary lighting) under specific microclimate conditions (such as specific temperature, humidity, and light conditions), assuming that the total energy demand over a period of time is E total , and calculate the total energy demand for irrigation, cooling and lighting (E irrigation 、E cooling 、E lig h ting );

[0121] Calculate the average proportion of each demand type: According to the average proportion of each demand type in the total demand in the historical data, calculate the weight of each demand, expressed as:

[0122]

[0123] S232, result output: different microclimate characteristics are input through the microclimate-driven energy demand model, and the energy demand prediction values ​​of irrigation, cooling and supplementary lighting under different microclimate conditions are output;

[0124] Through the above content, based on different microclimate characteristics, the microclimate-driven energy demand model can be used to provide accurate energy demand forecasts for irrigation, cooling and supplementary lighting in agricultural parks. By inputting different microclimate conditions, the corresponding energy demand can be quickly obtained, allowing the park to flexibly adjust resource allocation to meet dynamic demand changes in actual applications.

[0125] The land degradation feature extraction in S31 includes:

[0126] S311, calculation of soil salinization index: According to the concentration of soluble salt in the soil, the soil salinization index SI is calculated to reflect the impact of salinization on land productivity, expressed as:

[0127]

[0128] Where, m is the number of sampling points, EC i represents the conductivity value of the i-th sampling point;

[0129] S312, erosion level calculation: According to the volume loss rate of soil erosion, the erosion level EL is determined, which is expressed as:

[0130]

[0131] Among them, ER i is the erosion rate of the i-th sampling point, EL is the erosion level, which indicates the degree of soil erosion. A higher erosion level means that the land needs more irrigation and protection measures, which in turn increases energy demand;

[0132] Through the above content, the potential impact of land degradation on energy demand is quantified, which not only provides a scientific indicator of the degree of land degradation, but also enables agricultural parks to identify high-energy consumption areas in advance and optimize the allocation of resources such as irrigation and soil improvement. Through this refined feature extraction, park managers can make more accurate energy planning, reduce resource waste, support the sustainable development of agricultural parks, and at the same time reduce the long-term management costs caused by land degradation.

[0133] The additional energy demand model for degraded land in S32 adopts a Bayesian regression model, which includes:

[0134] S321, introduce microclimate data: Based on the land degradation characteristics, temperature and humidity are combined as additional variables to construct a Bayesian regression model, which is expressed as:

[0135] E=θ0+θ1SI+θ2EL+θ3T+θ4H+∈;

[0136] Where E is the additional energy demand, SI is the soil salinization index, EL is the erosion level, T and H are temperature and humidity respectively, θ0, θ1, θ2, θ3, θ4 are model parameters, indicating the weight of each variable on energy demand, ∈ is the noise term, which obeys the normal distribution, that is,

[0137] S322, introduce prior distribution: according to the different importance of land degradation characteristics and additional variables, different prior distributions are set. The prior distributions of θ1 and θ2 are biased towards larger values, reflecting the significant impact of land degradation characteristics on energy demand. The prior distributions of θ3 and θ4 are biased towards smaller values ​​to reflect the relatively minor impact of microclimate characteristics, expressed as:

[0138]

[0139] Among them, μ high is set to a higher mean, indicating the importance of land degradation characteristics, Set to a smaller variance to reflect the certainty of land degradation characteristics, μ low , have lower means and larger variances, respectively, to reflect the secondary effects and uncertainties of microclimate characteristics on energy demand;

[0140] S323, establish a hierarchical Bayesian model: Consider the differences between different plots, and use the hierarchical model to estimate the Bayesian model parameters of each plot. Suppose the additional energy demand of plot i is E i , the hierarchical model is expressed as:

[0141] E i =θ 0i +θ1i SI i +θ 2i EL i +θ 3i T i +θ 4i H i +∈ i ;

[0142] Among them, θ 0i ,θ 1i ,θ 2i ,θ 3i ,θ 4i Follows a normal distribution with population parameter, θ 0i ,θ 1i ,θ 2i ,θ 3i , θ0 represents the mean of the population parameter, τ 2 It represents the variance of parameters between plots, which is used to reflect the differences between different plots;

[0143] S324, update parameters and predict through posterior distribution: update parameters using Bayesian theorem and calculate posterior distribution, expressed as:

[0144] P(θ|E,X)∝P(E|X,θ)·P(θ);

[0145] Among them, P(θ|E,X) is the posterior distribution, which is used to update the understanding of parameter θ after observing data E and X, P(E|X,θ) is the likelihood function, which represents the probability of observing data under the condition of given θ, and P(θ) is the prior distribution;

[0146] S325, Future Energy Demand Forecast: Use the expected value of the posterior distribution to predict future energy demand, expressed as:

[0147]

[0148] in, is the predicted value of energy demand at the future moment, G(θ|E,X) is the expected value of the posterior distribution of the parameter θ of the Bayesian regression model after given data E and X, and X is the observed characteristic data, including land degradation characteristics, temperature and humidity;

[0149] Through the above content, land degradation characteristics (salinization index and erosion level) and microclimate data (temperature and humidity) are combined, and the differences between different plots are processed through a hierarchical structure. When there is less data or uncertainty in the characteristic variables, prior information and posterior updates can be used to improve the stability and accuracy of the prediction. At the same time, the Bayesian method makes the model parameters have clear confidence intervals, which is convenient for explaining the impact of characteristics on energy demand, thereby providing agricultural parks with more scientific and flexible energy management solutions, improving resource utilization efficiency, and supporting sustainable development.

[0150] The relationship mapping between land degradation and energy consumption in S33 includes:

[0151] S331, Relationship mapping analysis: Based on the Bayesian regression model, the energy demand at each degradation level is calculated;

[0152] S332, relationship diagram output: a curve showing the change of energy demand with the degree of land degradation is drawn through the result of relationship mapping analysis, and the generated relationship diagram between land degradation and energy demand is provided to the agricultural park manager;

[0153] Through the above content, the relationship between the degree of land degradation and energy demand can be clearly displayed in a graphical way through quantitative analysis. The energy demand curves under different degradation levels provide an intuitive description of the impact of land degradation on energy consumption, helping park managers to quickly identify high-energy consumption areas. It not only improves the scientific nature of energy management, but also supports the precise allocation of resources, reduces waste, optimizes the sustainable management of the park, and provides decision makers with effective tools and data basis.

[0154] The comprehensive energy risk assessment in S4 includes:

[0155] S41, Risk index calculation: Based on energy consumption data, microclimate conditions and land degradation degree, calculate the comprehensive energy risk index ERI, expressed as:

[0156]

[0157] Among them, E actual is the actual energy consumption, E predicted is the energy demand forecast value based on the microclimate-driven energy demand model, E degraded is the energy consumption of degraded land, E normal is the energy consumption of normal land, V microclimate is the fluctuation degree of the current microclimate variables, V baseline is the degree of fluctuation of the benchmark microclimate, w1, w2, and w3 are weight coefficients used to balance the impact of various indicators;

[0158] S42. Risk level determination: Based on the calculated risk index ERI, compare it with the preset upper risk threshold T2 and lower risk threshold T1. When ERI < T1, it indicates low risk; when T1 ≤ ERI < T2, it indicates medium risk; when ERI ≥ T2, it indicates high risk.

[0159] The upper risk threshold T2 and the lower risk threshold T1 are set based on the statistical analysis of historical data, specifically including:

[0160] Collect historical data: Collect the historical energy consumption data of the park under different microclimate conditions and land degradation degrees, and calculate the energy risk index ERI for each time period.

[0161] Determine the distribution interval: Calculate the mean μ ERI and the standard deviation σ ERI of the energy risk index ERI in the historical data. Assume that the energy risk index of the historical data follows a normal distribution, and use the mean and standard deviation to set reasonable risk thresholds.

[0162] Set the risk thresholds: Set the lower risk threshold T1 as the mean plus one standard deviation, that is, T1 = μ ERI + σ ERI Set the upper risk threshold T2 as the mean plus two standard deviations, that is, T2 = μ ERI + 2σ ERI ;

[0163] S43. Generate a risk warning signal: When medium risk or high risk is reached, generate a risk warning signal and send a warning message to the park manager.

[0164] Through the above content, combined with historical energy consumption data, microclimate changes, and land degradation degrees, quantifying energy risks with the dynamically calculated energy risk index can provide forward-looking risk warnings for the park, help managers identify potential energy problems early, allocate resources reasonably, ensure the stability and sustainability of energy management, and improve the operation efficiency and risk resistance ability of the park.

[0165] This invention covers any substitutions, modifications, equivalent methods, and solutions made within the essence and scope of this invention. To enable the public to have a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments of this invention. However, those skilled in the art can fully understand this invention without the description of these details. Additionally, to avoid unnecessary confusion to the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0166] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An energy risk assessment method for an integrated energy system in an agricultural park, characterized in that: The following steps are involved: S1, data collection and preprocessing: collect multidimensional data required for energy risk assessment of the integrated energy system in the agricultural park, including energy consumption data, microclimate data, and land degradation data, and preprocess the multidimensional data, including data cleaning and standardization; S2, Microclimate-driven energy demand forecast: Based on the collected microclimate data, the dynamic impact of microclimate change on the energy demand for irrigation, cooling and supplementary lighting is analyzed to form energy demand forecasts under different climate conditions; S3, Energy consumption feedback of land degradation: Based on the collected land degradation data, analyze the additional energy demand caused by land degradation and provide the relationship between the degree of land degradation and energy consumption, including: S31, land degradation feature extraction: based on the collected land degradation data, extract land degradation features, including soil salinization index and erosion grade; S32, Modeling of additional energy demand for degraded land: Based on the extracted land degradation characteristics, an additional energy demand model for degraded land is constructed; S33, Relationship mapping between land degradation and energy consumption: Through the additional energy demand model of degraded land, the relationship mapping between land degradation degree and energy consumption is analyzed, and the impact curve of different land degradation levels on energy demand is drawn to provide a relationship diagram between land degradation and energy consumption; S4, comprehensive energy risk assessment: Based on the pre-processed energy consumption data, combined with the results of microclimate-driven energy demand forecast and energy consumption feedback from land degradation, a comprehensive assessment of energy risks is conducted and risk warning signals are generated.

2. The energy risk assessment method for an agricultural park integrated energy system according to claim 1 is characterized in that: The data collection and preprocessing in S1 include: S11, Energy consumption data collection: Through smart electricity meters, smart gas meters, and flow meters, various energy consumption data in the agricultural park are collected in real time, including electricity, gas, and water consumption data, forming hourly or daily energy consumption records; S12, microclimate data collection: deploy temperature, humidity, wind speed, and light environment sensors in the agricultural park to collect microclimate data, including temperature, humidity, wind speed, and light; S13, land degradation data collection: use soil sensors and remote sensing imaging technology to obtain soil salinization and erosion data, and combine with geographic information systems to divide plots and collect land degradation data in different areas; S14, data cleaning: use outlier detection algorithm to identify and remove abnormal data points, normalize multidimensional data, and scale all data values ​​to the [0,1] interval.

3. The energy risk assessment method for an agricultural park integrated energy system according to claim 1 is characterized in that: The microclimate-driven energy demand forecast in S2 includes: S21, microclimate feature extraction: based on the collected microclimate data, extract key climate features related to energy demand, including cumulative temperature, humidity index, sunshine hours, and wind speed change rate; S22, microclimate-driven energy demand model construction: Based on the extracted key climate features, a microclimate-driven energy demand model is constructed; S23, demand output: The energy demand for irrigation, cooling and lighting under different microclimate conditions is analyzed through the microclimate-driven energy demand model.

4. The energy risk assessment method for an agricultural park integrated energy system according to claim 3 is characterized in that: The microclimate feature extraction in S21 includes: S211, cumulative temperature: by accumulating the daily average temperature, the cumulative temperature within a certain period of time is calculated to reflect the heat demand for crop growth; S212, Humidity Index: Combine relative humidity and air temperature to calculate the humidity index to reflect the impact of crops on irrigation needs; S213, sunshine hours: The sunshine hours are recorded by the weather station, reflecting the effectiveness of light, and recorded in hours; S214, wind speed change rate: calculates the degree of change of wind speed within a specified time period, reflecting the impact of wind speed on crop evaporation and cooling demand.

5. The energy risk assessment method for an agricultural park integrated energy system according to claim 4 is characterized in that: The microclimate-driven energy demand model in S22 adopts an autoregressive integrated moving average model, and the autoregressive integrated moving average model includes: S221, Feature selection and modeling objectives: Select microclimate characteristics that affect energy demand as exogenous variables and construct an autoregressive integrated moving average model; S222, difference processing: performing difference processing on the time series data to eliminate non-stationarity; S223, model parameter identification: using the autocorrelation function and the partial autocorrelation function to determine the order of the model; S224, Model fitting and validation: Use the least squares method to estimate parameters and perform model validation to ensure goodness of fit; S225, prediction and result output: use the fitted model to predict future energy demand and output the results.

6. The energy risk assessment method for an agricultural park integrated energy system according to claim 5 is characterized in that: The demand output in S23 includes: S231, Demand Decomposition: Based on Forecasted Energy Demand Allocate according to different demand types; S232, result output: Different microclimate characteristics are input through the microclimate-driven energy demand model, and the predicted values ​​of energy demand for irrigation, cooling and supplementary lighting under different microclimate conditions are output.

7. The energy risk assessment method for an agricultural park integrated energy system according to claim 1 is characterized in that: The land degradation feature extraction in S31 includes: S311, calculation of soil salinization index: According to the concentration of soluble salt in the soil, the soil salinization index SI is calculated to reflect the impact of salinization on land productivity; S312, erosion level calculation: Determine the erosion level EL based on the volume loss rate of soil erosion.

8. The energy risk assessment method for an agricultural park integrated energy system according to claim 7 is characterized in that: The additional energy demand model of the degraded land in S32 adopts a Bayesian regression model, and the Bayesian regression model includes: S321, introducing microclimate data: Based on the land degradation characteristics, temperature and humidity are combined as additional variables to build a Bayesian regression model; S322, introduction of prior distribution: different prior distributions are set according to the importance of land degradation characteristics and additional variables; S323, establishing a hierarchical Bayesian model: considering the differences between different plots, the Bayesian model parameters of each plot are estimated through a hierarchical model; S324, parameter update and prediction through posterior distribution: update the parameters using Bayesian theorem and calculate the posterior distribution; S325, Future Energy Demand Forecasting: Using the expected value of the posterior distribution to predict future energy demand.

9. The energy risk assessment method for an agricultural park integrated energy system according to claim 8 is characterized in that: The relationship mapping between land degradation and energy consumption in S33 includes: S331, Relationship mapping analysis: Based on the Bayesian regression model, the energy demand at each degradation level is calculated; S332, Relationship diagram output: Draw a curve of energy demand varying with the degree of land degradation based on the results of relationship mapping analysis, and provide the generated relationship diagram of land degradation and energy demand to the agricultural park manager.

10. The energy risk assessment method for an agricultural park integrated energy system according to claim 1, characterized in that: The comprehensive energy risk assessment in S4 includes: S41, Risk index calculation: Calculate the comprehensive energy risk index ERI based on energy consumption data, microclimate conditions, and the degree of land degradation. S42, Risk level determination: Compare the calculated risk index ERI with the preset upper risk threshold T2 and lower risk threshold T1. When ERI < T1, it indicates low risk; when T1 ≤ ERI < T2, it indicates medium risk; when ERI ≥ T2, it indicates high risk. S43, Generate a risk warning signal: When medium risk or high risk is reached, generate a risk warning signal and send a warning message to the park manager.

Images