CN-121983963-A - Light-storage integrated electric energy storage regulation and control system and method

CN121983963ACN 121983963 ACN121983963 ACN 121983963ACN-121983963-A

Abstract

The invention discloses a light-storage integrated electric energy storage regulation and control system and a method, and particularly relates to the technical field of electric energy storage regulation and control; according to the invention, the photovoltaic power generation rate and the load demand power prediction result are dynamically optimized by adopting rolling prediction logic and combining the power generation and load trend coefficients, the adjustment coefficients and the mapping rules based on multidimensional characteristic parameters of the photovoltaic side, the energy storage side, the load side and the power grid side, the traditional single-dimension prediction defect is overcome, the coupling influence of component temperature, illumination intensity and the like is fully considered, the prediction deviation is greatly reduced, and reliable data support is provided for accurate regulation.

Inventors

WANG XULEI
YU LONGZE
WANG CHUNMEI
DING WEI

Assignees

天津轻工职业技术学院

Dates

Publication Date: 20260505
Application Date: 20260203

Claims (10)

1. An optical storage integrated electric energy storage regulation and control system is characterized by comprising the following modules: The prediction evaluation module is used for rolling and predicting a photovoltaic power generation rate prediction sequence and a load demand power prediction sequence in the next regulation period based on the multidimensional characteristic parameters; The optimization decision module comprises a feature construction unit, a case retrieval unit and an adaptation correction unit; The characteristic construction unit is used for constructing a standardized scene characteristic vector by combining the obtained photovoltaic power generation rate prediction sequence and the load demand power prediction sequence with characteristic parameters ; The case searching unit searches the historical strategy case library, and each case in the case library is a triplet All the weighted Euclidean distance calculation libraries are adopted in the retrieval process Searching the first z historical cases with highest similarity coefficients as candidate strategies; The adaptation correction unit extracts a regulation strategy sequence and an effect evaluation vector of the candidate strategy, performs weighted fusion by taking a similarity coefficient as a reference, and screens the base strategy after determining a base evaluation coefficient; And the instruction issuing module takes the base strategy as a final regulation and control plan in the next regulation and control period.
2. The integrated optical storage and electrical energy storage conditioning system of claim 1, wherein: features contained in the multi-dimensional feature parameters; Photovoltaic side parameters, namely component temperature, historical power generation rate sequence and forecast illumination intensity; the energy storage side parameters are the current state of charge, temperature and charge and discharge efficiency of the battery; Load side parameters, namely real-time load power and adjustable load state; and the power grid side parameters are real-time-sharing electricity price information, required electricity charge information, and real-time frequency and voltage of the power grid.
3. The integrated optical storage and electrical energy storage conditioning system of claim 2, wherein: rolling prediction logic of a photovoltaic power generation rate prediction sequence; Photovoltaic power generation rate prediction sequence ; Wherein k is the total number of predicted time points in the next regulation period; The predicted power generation rate at the kth moment of the next regulation period is represented; Taking the photovoltaic power generation rate of each time point in the current time zone as a historical power generation rate sequence, extracting the characteristics of the historical power generation rate sequence, and calculating a power generation trend coefficient by adopting a linear fitting slope; predicting the photovoltaic power generation rate at each time point in the next regulation and control period by using the power generation trend coefficient, and taking the predicted photovoltaic power generation rate as an initial prediction sequence after the prediction is completed; extracting component temperatures at all times in a current time zone from multi-dimensional characteristic parameters, determining an adjustment coefficient by combining with the predicted illumination intensity, matching the adjustment coefficient with a preset adjustment coefficient reference range, and setting adjustment factors respectively corresponding to different matching results, wherein the matching results comprise a reference range, a higher reference range and a lower reference range; multiplying each group of predicted photovoltaic power generation rates in the initial predicted sequence by the adjustment factors to obtain a photovoltaic power generation rate predicted sequence 。
4. A light and storage integrated electrical energy storage regulation system as claimed in claim 3 wherein: Constructing logic of an initial prediction sequence; After the power generation trend coefficient is converted into an optimization coefficient by using a mapping rule, taking the last photovoltaic power generation rate in the current time zone as a reference value, and taking the result of the reference value multiplied by the optimization coefficient as the predicted photovoltaic power generation rate of the starting time point in the next regulation and control period, namely ; After predicting the predicted photovoltaic power generation rate of the time point in the next regulation period each time, updating the historical power generation rate sequence by using the predicted value, recalculating the power generation trend coefficient, and predicting the photovoltaic power generation rate of the next point; the photovoltaic power generation rate of each time point in the next regulation and control period is completely predicted and then is used as an initial prediction sequence; calculation logic for adjusting the coefficients; The method comprises the steps of calculating the average value of the component temperature at each moment to serve as the correction temperature of the power generation power in the next regulation and control period, taking the illumination intensity and the component temperature set under the standard test condition as the reference intensity and the reference temperature, carrying out weighted fusion on the forecast illumination intensity and the correction temperature in the current time zone in combination with the standard test condition, and outputting the adjustment coefficient of the initial prediction sequence of the next regulation and control period.
5. The integrated optical storage and management system according to claim 4, wherein: rolling prediction logic of the load demand power prediction sequence; Load demand power prediction sequence ; The predicted load demand power at the kth moment of the next regulation period is represented; Taking the real-time load power of each time point in the current time zone as a real-time sequence, extracting the characteristics of the real-time sequence, and calculating a load trend coefficient by adopting a linear fitting slope; After the load trend coefficient is converted into the load correction coefficient by mapping, the real-time load power at the last moment in the current time zone is taken as a basic value, and the result of the basic value multiplied by the load correction coefficient is taken as the predicted load demand power at the initial time point in the next regulation period, namely ; After predicting the predicted load demand power of the time point in the next regulation period each time, updating the real-time sequence by using the predicted value, recalculating the load trend coefficient, and predicting the predicted load demand power of the next point; After the point-by-point iteration is completed to complete the prediction of k time points, a load demand power prediction sequence is generated 。
6. The integrated optical storage and management system according to claim 5, wherein: Scene feature vector Constructing logic; Is a predictive feature; for the purpose of the energy storage feature, Representing an energy storage state of charge; is a charge-discharge power limit; Is the battery health; As a feature of the power grid, Representing upper and lower limits of grid-connected power; the electricity price difference is the peak-valley electricity price difference; to regulate the date type characteristic of the cycle, Respectively, a weekday, a weekend, and a holiday.
7. The integrated optical storage and management system according to claim 6, wherein: meaning description of triples; historical scene feature vectors before case execution; The strategy comprises the planned charge and discharge power and the operation mode of the energy storage system at each time interval in the next regulation period; The actual effect evaluation vector after executing the regulation strategy sequence comprises comprehensive benefits, load electricity shortage rate, power fluctuation stabilizing rate, energy storage life loss, energy storage charge-discharge duty ratio and grid-connected power violation times.
8. The integrated optical storage and management system of claim 7, wherein: Calculating a similarity coefficient; Scene feature vector In (a) and (b) Matching with each case of the historical strategy case library, and screening out The same case is used as a case to be evaluated; For the case to be evaluated Removing Then, carrying out standardization processing on the residual characteristic components; Respectively to using Euclidean distance algorithm 、 And Weighting and fusing after the distance calculation to obtain all Is a coefficient of similarity of (c).
9. The integrated optical storage and management system of claim 7, wherein: screening specific processes of the base strategy; From candidate strategies Extracting comprehensive benefits, load electricity shortage rate, power fluctuation stabilizing rate, energy storage life loss, energy storage charge-discharge duty ratio and grid-connected power violation times; The comprehensive income, the load electricity shortage rate, the power fluctuation stabilizing rate, the energy storage life loss, the energy storage charge-discharge duty ratio and the grid-connected power violation times are subjected to standardized treatment, and then weighted fusion is carried out to obtain an effect evaluation coefficient; And based on the similarity coefficient and the effect evaluation coefficient of the candidate strategies, obtaining the base evaluation coefficient of the candidate strategies after weighted fusion processing, and selecting the candidate strategy with the highest base evaluation coefficient as the base strategy.
10. A light-storage integrated electric energy storage regulation method applied to the light-storage integrated electric energy storage regulation system as set forth in any one of claims 1 to 9, characterized by comprising: State sensing, namely acquiring multidimensional characteristic parameters based on a current time zone, wherein the multidimensional characteristic parameters comprise photovoltaic side parameters, energy storage side parameters, load side parameters and power grid side parameters; The method comprises the steps of predicting and evaluating, based on multidimensional characteristic parameters, a photovoltaic power generation rate prediction sequence and a load demand power prediction sequence in the next regulation period in a rolling mode; optimizing the decision, constructing a standardized scene feature vector, searching a historical strategy case library to obtain candidate strategies, and fusing similarity coefficients and effect evaluation coefficients of the candidate strategies to screen a base strategy; and issuing instructions and updating cases, and issuing and executing the base strategy as a final regulation and control plan.

Description

Light-storage integrated electric energy storage regulation and control system and method Technical Field The invention relates to the technical field of electric energy storage regulation and control, in particular to an optical storage integrated electric energy storage regulation and control system and method. Background Along with acceleration of global energy transformation process, photovoltaic energy is used as clean and renewable high-quality energy, the duty ratio in an energy supply system is continuously improved, and the photovoltaic power generation and energy storage technology is organically combined by the photovoltaic integrated system, so that the impact of intermittence and fluctuation of photovoltaic output on stable operation of a power grid is effectively relieved, and the photovoltaic integrated system becomes a key support technology for promoting large-scale application of the photovoltaic energy and is widely applied to multiple scenes such as industrial production, commercial buildings, resident distributed energy and the like. However, the light storage integrated electric energy storage regulation and control system in the prior art has the following technical bottlenecks in the practical application process: firstly, the prediction precision is insufficient, the traditional regulation and control system relies on historical data with single dimension to predict photovoltaic power generation capacity and load demand, coupling influence of multidimensional characteristic parameters such as component temperature, illumination intensity change and the like is not fully considered, so that the deviation between a prediction result and an actual condition is large, and the accurate regulation and control decision is difficult to support; Secondly, the strategy adaptability is poor, the regulation and control strategies of the existing system are mostly fixed rules or offline optimization schemes, and cannot be dynamically adjusted according to real-time scenes, and when facing different date types such as workdays, weekends, holidays and the like, and complex scenes such as peak-valley electricity price fluctuation, power grid running state change and the like, the balance of regulation and control effects is difficult to realize; Thirdly, decision efficiency and effect are difficult to consider, and although part of systems introduce historical case references, a scientific similarity evaluation method and a multi-objective effect evaluation mechanism are lacked, so that an optimal strategy adapting to the current scene cannot be quickly screened out. Therefore, the system and the method for regulating and controlling the light storage integrated electric energy storage are provided. Disclosure of Invention In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide a light-storage integrated electrical energy storage regulation system and method. In order to achieve the above purpose, the present invention provides the following technical solutions: an optical storage integrated electric energy storage regulation and control system comprises the following modules: The prediction evaluation module is used for rolling and predicting a photovoltaic power generation rate prediction sequence and a load demand power prediction sequence in the next regulation period based on the multidimensional characteristic parameters; The optimization decision module comprises a feature construction unit, a case retrieval unit and an adaptation correction unit; The characteristic construction unit is used for constructing a standardized scene characteristic vector by combining the obtained photovoltaic power generation rate prediction sequence and the load demand power prediction sequence with characteristic parameters ; The case searching unit searches the historical strategy case library, and each case in the case library is a tripletAll the weighted Euclidean distance calculation libraries are adopted in the retrieval processSearching the first z historical cases with highest similarity coefficients as candidate strategies; The adaptation correction unit extracts a regulation strategy sequence and an effect evaluation vector of the candidate strategy, performs weighted fusion by taking a similarity coefficient as a reference, and screens the base strategy after determining a base evaluation coefficient; And the instruction issuing module takes the base strategy as a final regulation and control plan in the next regulation and control period. Specifically, the multidimensional feature parameters comprise features; Photovoltaic side parameters, namely component temperature, historical power generation rate sequence and forecast illumination intensity; the energy storage side parameters are the current state of charge, temperature and charge and discharge efficiency of the battery; Load side parameters, namely real-time load power and adjustable lo