CN-122022226-A - Urban level climate toughness assessment and distributed energy scheduling decision system

CN122022226ACN 122022226 ACN122022226 ACN 122022226ACN-122022226-A

Abstract

The invention discloses a city-level climate toughness assessment and distributed energy scheduling decision system, which relates to the technical field of assessment energy scheduling and is used for solving the problems that assessment results are lagged and lack of sensitivity, misjudgment phenomena such as regional boundary fracture or clustering offset are generated, and the energy scheduling logic is single, assessment model responsiveness and division accuracy are assessed through historical data weight change rate and clustering space dispersion, unstable regions are identified based on K-means, jenks and a fractional algorithm in a multiple classification mode, the unstable regions are identified by combining kappa coefficients, the risk change frequency is counted, the substantial unstable regions are identified by combining logistic regression, the misjudgment type is identified by utilizing a support vector machine, energy compression and elastic sharing are implemented on the false positive regions, rapid energy storage and cross-region scheduling are implemented on the false negative regions, the assessment accuracy and scheduling robustness are improved, the regional division accuracy and reliability are enhanced, and the intelligent and differential response of the energy scheduling are realized.

Inventors

TANG LINXIANG
LI XINYAO
Peng Daicheng
ZHANG YIXUAN

Assignees

东北大学秦皇岛分校

Dates

Publication Date: 20260512
Application Date: 20251205

Claims (10)

1. A city level climate toughness evaluation and distributed energy scheduling decision system is characterized by comprising a zoning analysis module, a classification evaluation module, a risk identification module and a false scheduling module, wherein the modules are connected by signals: The regional analysis module extracts the duty ratio weights of various historical data which participate in modeling of the urban level climate toughness prediction model in the historical database, performs ratio analysis on the duty ratio weights of the latest historical data, acquires the relative weight change rate, and analyzes the similar space discrete maximum value in the clustering algorithm in the model; The classification evaluation module comprehensively analyzes the relative weight change rate and the similar space discrete maximum value to judge whether the region division misjudgment risk exists, and if so, three algorithms are called to repeatedly classify the region, the consistency among classification results is compared, the classification consistency coefficient is calculated, and whether an unstable region exists is identified; The risk identification module screens an unstable region identified by the classification consistency coefficient, counts the accumulated number of risk cross-grade change and the risk grade conversion cycle frequency thereof, comprehensively analyzes the unstable region, and judges whether the region is a substantial unstable region or not and marks the region; the false scheduling module builds a false judgment probability index model to identify the false judgment type of the unstable region, and resource recovery or scheduling reinforcement measures are respectively adopted according to the false judgment type.
2. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 1, wherein: Collecting the latest historical data from a historical database, extracting a corresponding weight value of the latest historical data in the urban level climate toughness prediction model, and obtaining a relative weight change rate by the ratio of the latest historical data to the total historical data weight value; and calculating the square of Euclidean distance from the space unit in each clustering class to the gravity center of the class in the urban level climate toughness prediction model, taking the average value, and taking the maximum value to obtain the similar space discrete maximum value.
3. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 2, wherein: comparing the relative weight change rate with a weight sensitivity threshold, comparing the similar spatial discrete maximum value with a spatial dispersion threshold, and determining that the region division misjudgment risk exists if the relative weight change rate is smaller than the weight sensitivity threshold and the similar spatial discrete maximum value is larger than the spatial dispersion threshold.
4. A city level climate toughness assessment and distributed energy scheduling decision system as claimed in claim 3, wherein: And respectively obtaining corresponding region division results for regions with region division misjudgment risks based on a clustering algorithm, a natural breakpoint classification algorithm and a quantile classification algorithm.
5. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 4, wherein: calculating the Council Kappa coefficients of any two groups of division results according to the corresponding region division results; forming a consistency matrix by combining consistency coefficients of all pairwise combinations and calculating the division consistency coefficient of each region; And comparing the partition consistency coefficient with a judgment threshold value, and judging the region with the region partition misjudgment risk as a potential unstable region if the partition consistency coefficient is smaller than the judgment threshold value.
6. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 5, wherein: Accumulating cross-grade change of the climate toughness grading label in the continuous two time periods in the history time of uniformly dividing the time period, and obtaining the accumulated times of the risk cross-grade change; And calculating the ratio of the accumulated number of risk cross-level change and the total time period number of historical time to obtain the risk level conversion period frequency.
7. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 6, wherein: Taking the accumulated number of risk cross-level change and the risk level conversion cycle frequency as input features, and comparing the probability of the logistic regression output area being a substantial unstable area with a set unstable judgment threshold; if the probability that the potential unstable region is a substantially unstable region is greater than the instability determination threshold, the potential unstable region is determined to be a substantially unstable region and is marked as an unstable region.
8. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 7, wherein: calculating an absolute value of an implicit risk intensity difference between the historical extreme event frequency of the unstable region and the climate toughness model evaluation level, and obtaining an extreme event response residual; calculating the ratio of the energy allocation intensity of the unstable area to the climate risk index to obtain the resource utilization-risk contradiction ratio; And taking the extreme event response residual error and the resource utilization-risk contradiction ratio as two-dimensional feature vectors to be input, carrying out joint judgment by using a support vector machine classification model, and judging the misjudgment type according to the output value of the decision function, wherein the misjudgment type comprises false positive or false negative.
9. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 8, wherein: For the false positive region, the resource recovery and fault tolerance compression operation is executed, and the method specifically comprises the steps of energy redundancy compression, energy elastic sharing and execution cycle elastic adjustment.
10. The urban level climate toughness assessment and distributed energy scheduling decision system according to claim 8, wherein: and for the false negative area, executing a quick response type energy strengthening mechanism, wherein the quick response type energy strengthening mechanism specifically comprises quick energy storage deployment, predictive front-end scheduling and cross-area cooperation.

Description

Urban level climate toughness assessment and distributed energy scheduling decision system Technical Field The invention relates to the technical field of energy scheduling evaluation, in particular to a city-level climate toughness evaluation and distributed energy scheduling decision system. Background As global climate change increases, the climate toughness of cities in the face of extreme weather events (e.g., high temperature, heavy rain, drought, etc.) has become one of the key indicators to assess the sustainable development ability of cities. The urban climate toughness assessment aims at identifying high-risk areas by integrating data of cities in multiple dimensions such as infrastructure, resource guarantee, ecosystem and emergency response, and providing scientific regional management and emergency resource allocation basis for governments, and currently, most urban climate toughness assessment relies on multi-source historical data and regional cluster analysis models to realize risk grading division of urban space units. At present, most models do not respond dynamically to the change trend of the latest data when being constructed, so that the model is delayed and lacks sensitivity when facing the abrupt change of a climate structure and the frequent occurrence of extreme events, and meanwhile, the partial region division algorithm possibly ignores the trend of strong correlation between adjacent regions in space under the condition of untreated space autocorrelation, thereby generating misjudgment phenomena such as 'region boundary fracture' or 'clustering deviation', and the like, and causing single energy scheduling logic. Therefore, a city level climate toughness assessment and distributed energy scheduling decision system is provided to cope with the problem of insufficient sensitivity and space consistency of the model. Disclosure of Invention In order to overcome the defects in the prior art, the embodiment of the invention provides a city-level climate toughness assessment and distributed energy scheduling decision system, which solves the problems in the background art by integrating a historical data dynamic weight identification, cluster failure diagnosis, multi-algorithm reclassification consistency comparison, regional instability accumulation analysis and misjudgment type driven resource scheduling optimization mechanism. In order to achieve the above purpose, the invention provides a system for urban climate toughness assessment and distributed energy scheduling decision, which comprises a zoning analysis module, a classification assessment module, a risk identification module and a false scheduling module, wherein the modules are connected by signals: The regional analysis module extracts the duty ratio weights of various historical data which participate in modeling of the urban level climate toughness prediction model in the historical database, performs ratio analysis on the duty ratio weights of the latest historical data, acquires the relative weight change rate, and analyzes the similar space discrete maximum value in the clustering algorithm in the model; The classification evaluation module comprehensively analyzes the relative weight change rate and the similar space discrete maximum value to judge whether the region division misjudgment risk exists, and if so, three algorithms are called to repeatedly classify the region, the consistency among classification results is compared, the classification consistency coefficient is calculated, and whether an unstable region exists is identified; The risk identification module screens an unstable region identified by the classification consistency coefficient, counts the accumulated number of risk cross-grade change and the risk grade conversion cycle frequency thereof, comprehensively analyzes the unstable region, and judges whether the region is a substantial unstable region or not and marks the region; the false scheduling module builds a false judgment probability index model to identify the false judgment type of the unstable region, and resource recovery or scheduling reinforcement measures are respectively adopted according to the false judgment type. In a preferred embodiment, the latest historical data is collected from a historical database, the corresponding weight value of the latest historical data in the urban level climate toughness prediction model is extracted, and the ratio of the weight value to the total historical data weight value is used for obtaining the relative weight change rate; and calculating the square of Euclidean distance from the space unit in each clustering class to the gravity center of the class in the urban level climate toughness prediction model, taking the average value, and taking the maximum value to obtain the similar space discrete maximum value. In a preferred embodiment, the relative weight change rate is compared to a weight sensitivity threshold and the similar spatial disper