Method and system for capturing comprehensive energy data of low-carbon park

Abstract

The invention discloses a low-carbon park comprehensive energy data capturing method and system, and the method comprises the steps: collecting low-carbon park data, and carrying out the preprocessing; and comprehensively evaluating the correlation and dependency between the preprocessed data and the overall energy consumption. And based on the correlation and dependency of the data and the overall energy consumption, establishing an energy data model, and analyzing a dynamic change relationship between different energy types and the overall energy consumption of the park. And generating an energy use optimization suggestion according to the dynamic change relationship. According to the method, the time sequence change and nonlinear dependency of different energy systems and the overall energy consumption are effectively captured, and more accurate inter-system mutual influence evaluation is realized. The correlation and dependency are introduced into energy consumption prediction and scheduling optimization, and the genetic algorithm is used to dynamically adjust the system weight, so that an intelligent energy scheduling scheme is provided, and the load response of the energy system is more flexible and accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of low-carbon park energy management, and in particular to a method and system for capturing comprehensive energy data of a low-carbon park. Background Art

[0002] With the global emphasis on carbon emission control, low-carbon parks have become one of the important ways to achieve sustainable development. In low-carbon parks, various energy systems (such as photovoltaic power generation, battery energy storage, heat pump systems, etc.) are widely used to provide electricity, cooling and heating. However, with the increase in equipment and energy types in the park, how to efficiently collect, integrate and analyze these energy data has become a technical challenge. At present, most energy management systems in the park rely on a single or decentralized energy data monitoring solution, which makes it difficult to comprehensively evaluate the synergy and dynamic changes of various energy systems. Most of the existing energy management methods use static data models, which can only perform statistical analysis on historical energy consumption. They lack real-time data capture and dynamic adjustment mechanisms, making it difficult to achieve efficient and accurate energy scheduling. In addition, existing technologies often cannot effectively analyze the interdependence between different energy systems, and fail to make full use of complex computational models such as information entropy to analyze system dependencies, resulting in the inability to dynamically adapt to the real-time needs of the park in energy optimization.

[0003] Existing technologies have deficiencies in many aspects. First, the collection and preprocessing process of energy data lacks flexibility and cannot cope with the diverse needs of various energy systems in the park, such as the real-time collection and cleaning of multi-dimensional data such as temperature, voltage, and air volume. Secondly, in correlation and dependency analysis, existing technologies usually use linear models, ignoring the nonlinear dynamic interactions between different systems. In addition, most of the optimization algorithms in existing technologies use traditional weighted models, fail to introduce intelligent heuristic algorithms, and cannot provide dynamic optimization suggestions for complex multi-system energy scheduling. These limitations have led to inefficient energy management, and it is difficult for the system to respond to changes in the energy consumption demand of the park in real time. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the present invention solves the problem that it is difficult to achieve real-time dynamic data analysis and energy scheduling optimization in the prior art. Most of the prior art relies on static historical data and cannot respond to the dynamic changes of various energy systems in the park in a timely manner. In addition, the linear optimization algorithm in traditional technology is difficult to cope with complex multi-system interactions and nonlinear dependencies. The present invention uses genetic algorithms for heuristic optimization, which effectively solves the problems of inaccurate system energy consumption prediction and rigid energy scheduling.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for capturing comprehensive energy data of a low-carbon park, comprising: collecting low-carbon park data and performing preprocessing.

[0007] Comprehensively evaluate the correlation and dependence of preprocessed data with overall energy consumption.

[0008] Based on the correlation and dependence between data and overall energy consumption, an energy data model is established to analyze the dynamic relationship between different energy types and the overall energy consumption of the park.

[0009] Generate energy usage optimization recommendations based on dynamically changing relationships.

[0010] As a preferred solution of the low-carbon park comprehensive energy data capture method described in the present invention, the collection of low-carbon park data and preprocessing includes collecting basic data type systems, photovoltaic systems, battery systems, heat pump systems, water storage systems, wind disks and lighting data within the park.

[0011] The basic data type system includes temperature.

[0012] Photovoltaic systems include current and temperature.

[0013] The battery system includes current, voltage, temperature, and remaining capacity.

[0014] The heat pump system includes the water temperature and flow rate on the primary and secondary sides.

[0015] The water storage system includes inlet and outlet water temperatures and stratified water temperatures.

[0016] The fan disk includes temperature and air volume.

[0017] Lighting data includes illuminance, brightness and power.

[0018] Preprocessing includes data cleaning and data normalization.

[0019] Preprocessing included imputing missing values ​​and data normalization.

[0020] The missing values ​​were filled using the mean imputation method, and the data standardization was expressed as:

[0021]

[0022] Among them, x i ′ represents the standardized data, x i represents the original data, x min Represents the minimum value in the data, x max Indicates the maximum value in the data.

[0023] As a preferred solution of the method for capturing comprehensive energy data of low-carbon parks described in the present invention, the correlation and dependence between the pre-processed data and the overall energy consumption of the comprehensive evaluation include: the correlation is used to measure the linear relationship between each energy system and the overall energy consumption, calculated using the Pearson correlation coefficient, expressed as:

[0024]

[0025] Among them, x i represents the energy data of the ith system, y i Represents the overall energy consumption data at time i, Represents the mean value of system energy data, Indicates the average value of overall energy consumption data.

[0026] The time-weighted dynamic correlation coefficient is introduced, and a decreasing exponential decay function is applied in the time dimension. The weighted term is expressed as:

[0027]

[0028] Where α represents the time attenuation coefficient, t n represents the current time, t i represents the i-th moment.

[0029] The Pearson correlation coefficient after introducing time decay is expressed as:

[0030]

[0031] The Pearson correlation coefficient after introducing time decay ranges from -1 to 1. A value close to 1 indicates a strong positive correlation between the energy system and the overall energy consumption, a value close to -1 indicates a strong negative correlation, and a value close to 0 indicates no correlation.

[0032] As a preferred solution of the low-carbon park comprehensive energy data capture method described in the present invention, the correlation and dependence between the pre-processed data and the overall energy consumption of the comprehensive evaluation also include that the dependence reflects the degree of influence of a certain system on the overall energy consumption, and whether the system can determine the change of overall energy consumption. The conditional entropy and mutual information based on information theory are used for calculation. The original conditional entropy formula is the uncertainty of system X under the premise of known system Y, which is expressed as:

[0033] H(X|Y(=H(X|Y)-H(Y)

[0034] Among them, H(X|Y) represents conditional entropy, which indicates the uncertainty of X when Y is known; H(X|Y) represents joint entropy, which indicates the common uncertainty of X and Y; and H(Y) represents the entropy of Y, which indicates the uncertainty of Y.

[0035] A nonlinear dependency index is introduced to enhance the capture of nonlinear dependencies by adding a mutual information filter function, which is expressed as:

[0036]

[0037] Among them, D(X,Y) represents the dependence of system X on the overall energy consumption Y, and I(X;Y) represents the mutual information between X and Y, capturing the information dependency between the two. i-1 ) represents the conditional entropy of the previous moment, which is used to construct the temporal dependency of the system. β represents the contribution of the control mutual information to the dependency, ensuring that the dependency is higher when the mutual information is larger. N represents the total number of sampling points.

[0038] The optimized dependency range is 0 to 1, where 0 means no dependency and 1 means full dependency. The closer the value is to 1, the greater the impact of the system on the overall energy consumption.

[0039] As a preferred solution of the low-carbon park comprehensive energy data capture method described in the present invention, the establishment of the energy data model includes constructing a comprehensive energy consumption model based on correlation and dependency, and analyzing the dynamic relationship between different energy types and the overall energy consumption of the park.

[0040] The overall energy consumption is calculated based on the correlation coefficient and the data of each energy system, expressed as:

[0041]

[0042] in, represents the predicted overall energy consumption at time t, m represents the number of energy systems, and w i Represents the weight of the i-th system, reflecting its impact on the overall energy consumption. dynamic,i (t) represents the dynamic correlation coefficient of the ith system at the tth moment, E i (t) represents the energy consumption data of the i-th system at the t-th moment.

[0043] In dependency analysis, the energy consumption contribution in the model is adjusted based on the dependency of each system on the overall energy consumption. Dependency is used to adjust the influence of each system and ultimately affect its weight. We perform nonlinear adjustment on the weight in energy consumption prediction, expressed as:

[0044]

[0045] Among them, w i ′ represents the weight of the ith system after dependency adjustment, D(X i ,Y) represents the dependence of the ith system on the overall energy consumption, w iRepresents the initial weight, which is the proportion of each system's impact on the overall energy consumption when not adjusted.

[0046] As a preferred solution of the low-carbon park comprehensive energy data capture method described in the present invention, the establishment of the energy data model also includes introducing a genetic algorithm to optimize the weight w and dependency parameters in the model to make the overall energy consumption prediction more accurate and provide the optimal energy scheduling solution.

[0047] Randomly generate a set of weights {w i} and the dependency {D(X i ,Y)}, each solution represents a different energy scheduling scheme.

[0048] Define the fitness function f(w 1 ,w 2 ,…,w m ,D 1 ,D 2 ,…,D m ) is the inverse of the energy consumption prediction error, and the goal is to minimize the prediction error, which can be expressed as:

[0049]

[0050] Among them, E true (t) represents the actual energy consumption value, Indicates the predicted energy consumption value.

[0051] Individuals with high fitness are selected as parents to enter the next generation.

[0052] Perform a crossover operation on the selected parent generation to generate new candidate solutions.

[0053] The weight w of some individuals that mutate randomly i and the dependency D(X i ,Y), introduce randomness to avoid local optimality.

[0054] Continue to iterate crossover and mutation until the fitness function converges.

[0055] The weight w optimized by genetic algorithm i ' and the dependency D(Xi,Y) are used in the final energy consumption prediction model to ensure optimal energy scheduling and energy consumption prediction.

[0056] Taking into account the correlation, dependency and weight adjustment after genetic algorithm optimization, the final energy consumption prediction is expressed as:

[0057]

[0058] Among them, w i ′ represents the weight after optimization by genetic algorithm, rdynamic,i (t) represents the dynamic correlation coefficient, E i (t) is the energy consumption data of each system.

[0059] As a preferred solution of the low-carbon park comprehensive energy data capture method described in the present invention, the generation of energy usage optimization suggestions based on dynamic change relationships includes formulating dynamic energy scheduling strategies based on the correlation and dependence between the energy system and the overall energy consumption, giving priority to scheduling systems that are highly correlated with the overall energy consumption and have a high degree of dependence, and reducing the use of inefficient energy.

[0060] During peak load periods, efficient and stable systems are dispatched, and during low load periods, renewable energy such as photovoltaic systems are used to supplement energy supply to optimize the overall dispatch of the energy system.

[0061] Through long-term dependency analysis, we can identify inefficient or unstable systems and make maintenance and upgrade recommendations. We regularly maintain inefficient systems to ensure their normal operation. For systems that have been inefficient for a long time, we can make hardware upgrade recommendations.

[0062] 8 A low-carbon park comprehensive energy data capture system, characterized by: including:

[0063] The preprocessing module collects low-carbon park data and performs preprocessing.

[0064] The evaluation module comprehensively evaluates the correlation and dependence between the preprocessed data and the overall energy consumption.

[0065] The analysis module establishes an energy data model based on the correlation and dependence between data and overall energy consumption, and analyzes the dynamic relationship between different energy types and the overall energy consumption of the park.

[0066] The suggestion module generates energy usage optimization suggestions based on the dynamically changing relationships.

[0067] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0068] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0069] Beneficial effects of the present invention: The present invention achieves optimized management of comprehensive energy in low-carbon parks through four steps. By comprehensively evaluating the correlation and dependence of each energy system, a dynamic correlation coefficient and a nonlinear dependency index based on time decay are proposed, which effectively captures the temporal changes and nonlinear dependencies of different energy systems and overall energy consumption, and achieves a more accurate assessment of the mutual impact between systems. By establishing an energy data model, introducing correlation and dependence into energy consumption prediction and scheduling optimization, and using genetic algorithms to dynamically adjust system weights, an intelligent energy scheduling solution is provided, making the load response of the energy system more flexible and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0071] Figure 1 An overall flow chart of a method for capturing comprehensive energy data for a low-carbon park provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0073] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a method for capturing comprehensive energy data of a low-carbon park, comprising:

[0074] S1: Collect low-carbon park data and perform preprocessing.

[0075] It should be noted that in low-carbon parks, in order to reduce carbon emissions, we must first actively adopt clean energy such as solar energy, wind energy and geothermal energy. These renewable energy sources do not produce carbon dioxide emissions themselves and can gradually replace traditional fossil energy power supply and heating, thereby providing clean electricity and heat for the daily operation of buildings. On this basis, we further collect and analyze data from various renewable energy systems (such as photovoltaic power generation, battery energy storage, ground source heat pumps, water storage cooling and heat storage systems, etc.). The data is shown in Table 1.

[0076] Table 1 Data and collection methods

[0077]

[0078]

[0079] Collect data on basic data types systems, photovoltaic systems, battery systems, heat pump systems, water storage systems, wind turbines and lighting within the park.

[0080] The basic data type system includes temperature.

[0081] Photovoltaic systems include current and temperature.

[0082] The battery system includes current, voltage, temperature, and remaining capacity.

[0083] The heat pump system includes the water temperature and flow rate on the primary and secondary sides.

[0084] The water storage system includes inlet and outlet water temperatures and stratified water temperatures.

[0085] The fan disk includes temperature and air volume.

[0086] Lighting data includes illuminance, brightness and power.

[0087] Preprocessing includes data cleaning and data normalization.

[0088] Preprocessing included imputing missing values ​​and data normalization.

[0089] The missing values ​​were filled using the mean imputation method, and the data standardization was expressed as:

[0090]

[0091] Among them, x i ′ represents the standardized data, x i represents the original data, x min Represents the minimum value in the data, x max Indicates the maximum value in the data.

[0092] S2: Comprehensively evaluate the correlation and dependence between the preprocessed data and the overall energy consumption.

[0093] Correlation is used to measure the linear relationship between each energy system and overall energy consumption, calculated using the Pearson correlation coefficient, expressed as:

[0094]

[0095] Among them, x i represents the energy data of the ith system, y i Represents the overall energy consumption data at time i, Represents the mean value of system energy data, Indicates the average value of overall energy consumption data.

[0096] The time-weighted dynamic correlation coefficient is introduced, and a decreasing exponential decay function is applied in the time dimension. The weighted term is expressed as:

[0097]

[0098] Where α represents the time attenuation coefficient, t n represents the current time, t i represents the i-th moment.

[0099] It should be noted that the time-weighted dynamic correlation coefficient is a method of optimization based on the Pearson correlation coefficient. The traditional Pearson correlation coefficient only considers the static linear relationship between variables and cannot reflect changes in the time dimension. In the energy system of the low-carbon park, the relationship between different energy systems and overall energy consumption is constantly changing over time. Therefore, the time-weighted dynamic correlation coefficient is introduced, and the exponential decay function is used to weight the historical data. Through the decreasing exponential decay function, higher weights are given to data closer to the current moment, while earlier data are given lower weights, thereby more accurately reflecting the real-time dynamic changes of the system.

[0100] Furthermore, the exponential decay function enables the model to give priority to the most recent data, effectively capture the dynamic correlation of energy systems over time, and adapt to the rapidly changing load demands of the park. By decaying older data, the interference of historical data on current analysis can be reduced, improving the real-time and accuracy of the system. The optimized dynamic correlation coefficient can better reflect the energy consumption contribution of each current energy system, help the dispatching system give priority to energy systems with strong correlation, and improve overall energy efficiency.

[0101] The Pearson correlation coefficient after introducing time decay is expressed as:

[0102]

[0103] The Pearson correlation coefficient after introducing time decay ranges from -1 to 1. A value close to 1 indicates a strong positive correlation between the energy system and the overall energy consumption, a value close to -1 indicates a strong negative correlation, and a value close to 0 indicates no correlation.

[0104] Dependence reflects the degree of influence of a certain system on the overall energy consumption, and whether the system can determine the change in overall energy consumption. The calculation is based on conditional entropy and mutual information based on information theory. The original conditional entropy formula is the uncertainty of system X under the premise of known system Y, expressed as:

[0105] H(X|Y(=H(X|Y)-H(Y)

[0106] Among them, H(X|Y) represents conditional entropy, which indicates the uncertainty of X when Y is known; H(X|Y) represents joint entropy, which indicates the common uncertainty of X and Y; and H(Y) represents the entropy of Y, which indicates the uncertainty of Y.

[0107] A nonlinear dependency index is introduced to enhance the capture of nonlinear dependencies by adding a mutual information filter function, which is expressed as:

[0108]

[0109] Among them, D(X,Y) represents the dependence of system X on the overall energy consumption Y, and I(X;Y) represents the mutual information between X and Y, capturing the information dependency between the two. i-1 ) represents the conditional entropy of the previous moment, which is used to construct the temporal dependency of the system. β represents the contribution of the control mutual information to the dependency, ensuring that the dependency is higher when the mutual information is larger. N represents the total number of sampling points.

[0110] The optimized dependency range is 0 to 1, where 0 means no dependency and 1 means full dependency. The closer the value is to 1, the greater the impact of the system on the overall energy consumption.

[0111] It should be noted that the core of optimizing the dependency is to build it through conditional entropy and mutual information based on information theory, combined with the time dimension and nonlinear dependencies between systems. On this basis, the mutual information filter function is introduced to enhance the ability to capture the nonlinear relationship between the system and the overall energy consumption. The optimized dependency can reflect the contribution of a system to the overall energy consumption in the current time period by adjusting the weight and time series relationship, rather than just a simple linear dependency.

[0112] Furthermore, the optimized dependency model can handle the complex interactions between multiple systems, especially nonlinear dependencies, which helps to more accurately evaluate the impact of the system on overall energy consumption. The introduced conditional entropy and mutual information combined with time series dependency can reflect the changes in the system's dependency at different times, allowing the system to dispatch energy more accurately. The optimized dependency can dynamically evaluate the contribution of the system, thereby adjusting energy use according to real-time needs and avoiding energy waste caused by over-reliance on a certain system.

[0113] S3: Based on the correlation and dependence between data and overall energy consumption, an energy data model is established to analyze the dynamic relationship between different energy types and the overall energy consumption of the park.

[0114] Based on correlation and dependence, a comprehensive energy consumption model is constructed to analyze the dynamic relationship between different energy types and the overall energy consumption of the park.

[0115] The overall energy consumption is calculated based on the correlation coefficient and the data of each energy system, expressed as:

[0116]

[0117] in, represents the predicted overall energy consumption at time t, m represents the number of energy systems, and w i Represents the weight of the i-th system, reflecting its impact on the overall energy consumption. dynamic,i (t) represents the dynamic correlation coefficient of the ith system at the tth moment, E i (t) represents the energy consumption data of the i-th system at the t-th moment.

[0118] In dependency analysis, the energy consumption contribution in the model is adjusted based on the dependency of each system on the overall energy consumption. Dependency is used to adjust the influence of each system and ultimately affect its weight. We perform nonlinear adjustment on the weight in energy consumption prediction, expressed as:

[0119]

[0120] Among them, w i ′ represents the weight of the ith system after dependency adjustment, D(X i ,Y) represents the dependence of the ith system on the overall energy consumption, w i Represents the initial weight, which is the proportion of each system's impact on the overall energy consumption when not adjusted.

[0121] It should be noted that the comprehensive energy consumption model dynamically predicts the overall energy consumption of the park by integrating the two indicators of correlation and dependence. The model not only considers the historical data of each system, but also adjusts the weight of the system in real time through dynamic correlation coefficients and dependencies. The adjustment of weights is based on the real-time calculated dependencies and correlations to ensure that the contribution of each energy system is consistent with its actual impact on the overall energy consumption, thereby optimizing energy consumption prediction and energy scheduling strategies.

[0122] Genetic algorithm is introduced to optimize the weight w and dependency parameters in the model, making the overall energy consumption prediction more accurate and providing the optimal energy scheduling solution.

[0123] Randomly generate a set of weights {w i} and the dependency {D(X i ,Y)}, each solution represents a different energy scheduling scheme.

[0124] Define the fitness function f(w 1 ,w 2 ,…,w m ,D 1 ,D 2 ,…,D m) is the inverse of the energy consumption prediction error, and the goal is to minimize the prediction error, which can be expressed as:

[0125]

[0126] Among them, E true (t) represents the actual energy consumption value, Indicates the predicted energy consumption value.

[0127] Individuals with high fitness are selected as parents to enter the next generation.

[0128] Perform a crossover operation on the selected parent generation to generate new candidate solutions.

[0129] The weight w of some individuals that mutate randomly i and the dependency D(X i ,Y), introduce randomness to avoid local optimality.

[0130] Continue to iterate crossover and mutation until the fitness function converges.

[0131] The weight w optimized by genetic algorithm i ' and the dependency D(Xi,Y) are used in the final energy consumption prediction model to ensure optimal energy scheduling and energy consumption prediction.

[0132] It should be noted that the introduction of heuristic algorithms is to solve the optimization problem in the complex scheduling of multi-energy systems. The traditional energy scheduling model is based on fixed weights and linear optimization, which is difficult to cope with the complex interactions and dynamic changes between multiple systems. Genetic algorithms randomly generate multiple candidate solutions and perform iterative optimization through operations such as selection, crossover, and mutation to gradually find the energy scheduling solution with the lowest energy consumption and the best performance. In genetic algorithms, the fitness function is used to measure the energy consumption performance of each candidate solution, thereby guiding the algorithm to converge toward the optimal solution.

[0133] Furthermore, heuristic algorithms can handle complex nonlinear and multivariable optimization problems, breaking through the limitations of traditional linear models, and are particularly suitable for complex energy scheduling problems between multiple systems. Through multiple generations of iterative optimization, heuristic algorithms can effectively avoid local optimal problems and find the global optimal scheduling solution, thereby minimizing the overall energy consumption of the park. Heuristic algorithms can flexibly adjust energy scheduling strategies according to different time periods and load requirements, thereby improving the overall energy management efficiency and flexibility of the park.

[0134] Taking into account the correlation, dependency and weight adjustment after genetic algorithm optimization, the final energy consumption prediction is expressed as:

[0135]

[0136] Among them, w i′ represents the weight after optimization by genetic algorithm, r dynamic,i (t) represents the dynamic correlation coefficient, E i (t) is the energy consumption data of each system.

[0137] S4: Generate energy usage optimization suggestions based on dynamic change relationships.

[0138] Based on the correlation and dependence between the energy system and the overall energy consumption, a dynamic energy scheduling strategy is formulated to give priority to scheduling systems that are highly correlated and highly dependent on the overall energy consumption, thereby reducing the use of inefficient energy.

[0139] During peak load periods, efficient and stable systems are dispatched, and during low load periods, renewable energy such as photovoltaic systems are used to supplement energy supply to optimize the overall dispatch of the energy system.

[0140] Through long-term dependency analysis, we can identify inefficient or unstable systems and make maintenance and upgrade recommendations. We regularly maintain inefficient systems to ensure their normal operation. For systems that have been inefficient for a long time, we can make hardware upgrade recommendations.

[0141] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for capturing comprehensive energy data of a low-carbon park is implemented.

[0142] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0143] Example 2, an embodiment of the present invention, provides a method and system for capturing comprehensive energy data of a low-carbon park. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0144] In this experiment, a comparative study was conducted on the comprehensive energy data capture method of low-carbon parks to verify the innovation and advantages of our invention compared with the existing technology. The experiment is divided into two parts, one using the existing energy management system and the other using the comprehensive energy data capture method of low-carbon parks invented by us. The test objects include photovoltaic systems, battery systems, heat pump systems, water storage systems and other energy systems in the park, and the parameters include temperature, current, voltage, energy consumption, etc.

[0145] Existing technologies mainly rely on independent energy data monitoring solutions for each system, and the collected data include the current and temperature of photovoltaic power generation, the voltage and remaining power of the battery system, the water temperature of the heat pump system, etc. After the data collection is completed, the energy consumption of the historical data is calculated through static analysis software.

[0146] Based on the low-carbon park comprehensive energy data capture method of the present invention, the experiment first obtains the data of photovoltaic, battery, heat pump and other systems through the multi-dimensional data acquisition module, and performs data preprocessing, including data cleaning, normalization and missing value filling. Subsequently, based on the Pearson correlation coefficient and the time-weighted dynamic correlation coefficient, the dynamic changes of each energy system and the overall energy consumption are evaluated. In terms of dependency calculation, the experiment uses conditional entropy combined with mutual information filtering function to capture the nonlinear dependencies between systems. In the final scheduling stage, a genetic algorithm is introduced to dynamically adjust the weights of each system to achieve priority scheduling of high-efficiency energy, thereby minimizing overall energy consumption and optimizing energy use. The experimental results are shown in Table 1.

[0147] Table 1 Experimental results

[0148]

[0149] By comparing the table data, we can clearly see the significant advantages of our invention. First, in terms of total energy consumption, the low-carbon park comprehensive energy data capture method invented by us can effectively reduce energy consumption. The energy consumption of the photovoltaic system and the battery system has been reduced by 12.5% ​​and 8.2% respectively. This is due to the time-weighted dynamic correlation coefficient and genetic algorithm optimization of the present invention, which makes energy scheduling more accurate. Secondly, in the comparison of system correlation and dependence, the prior art is only based on static analysis of historical data. The correlation coefficients of photovoltaic and battery systems are 0.65 and 0.55 respectively. Our invention uses time weighting and conditional entropy optimization, and the correlation is increased to 0.75 and 0.70. The dependence is also increased accordingly, reflecting the system's higher influence on overall energy consumption. This improvement makes energy scheduling more dynamic and can respond to changes in park load demand in real time.

[0150] Peak energy consumption data also demonstrates the advantages of our invention. In traditional systems, peak energy consumption usually increases significantly, while our invented scheduling strategy prioritizes the use of efficient systems and reduces the participation of inefficient systems, resulting in a reduction of peak energy consumption by about 10%. In addition, the improvement in scheduling flexibility scores shows that the invention significantly improves the real-time response capability of energy scheduling through heuristic optimization of genetic algorithms, and the scheduling response time is shortened from 120 seconds to 80 seconds.

[0151] The fixed dispatch strategy in the existing technology is difficult to cope with the load fluctuation of the park, resulting in a high system load fluctuation rate (20% for the battery system). However, our invention has significantly reduced the system load fluctuation rate to 8% through nonlinear dependency adjustment, ensuring the stability of energy supply. This stable load response greatly reduces equipment wear and sharp fluctuations in energy consumption peaks and valleys, further improving the energy efficiency and equipment life of the park.

[0152] In summary, the experimental data clearly demonstrates the innovation of our invention in energy management and its obvious advantages in practical applications. Through the optimized data model and scheduling algorithm, the present invention not only improves the synergy between systems, but also effectively reduces the overall energy consumption, and has significant economic benefits and environmental value.

[0153] Embodiment 3, an embodiment of the present invention provides a low-carbon park comprehensive energy data capture system, including a preprocessing module, which collects low-carbon park data and performs preprocessing. An evaluation module comprehensively evaluates the correlation and dependence between the preprocessed data and the overall energy consumption. An analysis module, based on the correlation and dependence between the data and the overall energy consumption, establishes an energy data model to analyze the dynamic relationship between different energy types and the overall energy consumption of the park. A suggestion module, generates energy usage optimization suggestions based on the dynamic relationship.

[0154] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for capturing comprehensive energy data of a low-carbon park, characterized in that: include: Collect low-carbon park data and perform pre-processing; Comprehensively evaluate the correlation and dependence between pre-processed data and overall energy consumption; Based on the correlation and dependence between data and overall energy consumption, an energy data model is established to analyze the dynamic relationship between different energy types and the overall energy consumption of the park; Generate energy usage optimization recommendations based on dynamically changing relationships.

2. The low-carbon park comprehensive energy data capture method according to claim 1, characterized in that: The collecting of low-carbon park data and preprocessing includes collecting data of basic data types systems, photovoltaic systems, battery systems, heat pump systems, water storage systems, wind turbines and lighting in the park; The basic data type system includes temperature; Photovoltaic systems include current and temperature; The battery system includes current, voltage, temperature, and remaining capacity; The heat pump system includes water temperature and flow on the primary and secondary sides; The water storage system includes inlet and outlet water temperatures and stratified water temperatures; The fan disk includes temperature and air volume; Lighting data includes illuminance, brightness, and power; Preprocessing includes data cleaning and data normalization; Preprocessing included imputing missing values ​​and data standardization; The missing values ​​were filled using the mean imputation method, and the data standardization was expressed as: Among them, x i ′ represents the standardized data, x i represents the original data, x min Represents the minimum value in the data, x max Indicates the maximum value in the data.

3. The method for capturing comprehensive energy data of a low-carbon park according to claim 2, characterized in that: The correlation and dependence between the pre-processed data and the overall energy consumption of the comprehensive evaluation include: the correlation is used to measure the linear relationship between each energy system and the overall energy consumption, which is calculated using the Pearson correlation coefficient and is expressed as: Among them, x i represents the energy data of the ith system, y i Represents the overall energy consumption data at time i, Represents the mean value of system energy data, Represents the average value of overall energy consumption data; The time-weighted dynamic correlation coefficient is introduced, and a decreasing exponential decay function is applied in the time dimension. The weighted term is expressed as: Where α represents the time attenuation coefficient, t n represents the current time, t i represents the i-th moment; The Pearson correlation coefficient after introducing time decay is expressed as: The Pearson correlation coefficient after introducing time decay ranges from -1 to 1. A value close to 1 indicates a strong positive correlation between the energy system and the overall energy consumption, a value close to -1 indicates a strong negative correlation, and a value close to 0 indicates no correlation.

4. The method for capturing comprehensive energy data of a low-carbon park according to claim 3, characterized in that: The correlation and dependence between the pre-processed data and the overall energy consumption of the comprehensive evaluation also include that the dependence reflects the degree of influence of a certain system on the overall energy consumption, and whether the system can determine the change of the overall energy consumption; the conditional entropy and mutual information based on information theory are used for calculation. The original conditional entropy formula is the uncertainty of system X under the premise of known system Y, which is expressed as: H(X|Y)=H(X|Y)-H(Y) Among them, H(X|Y) represents conditional entropy, which indicates the uncertainty of X when Y is known; H(X|Y) represents joint entropy, which indicates the common uncertainty of X and Y; H(Y) represents the entropy of Y, which indicates the uncertainty of Y; A nonlinear dependency index is introduced to enhance the capture of nonlinear dependencies by adding a mutual information filter function, which is expressed as: Among them, D(X,Y) represents the dependence of system X on the overall energy consumption Y, I(X;Y) represents the mutual information between X and Y, capturing the information dependency between the two; H(X|Y i-1 ) represents the conditional entropy of the previous moment, which is used to construct the temporal dependency of the system; β represents the contribution of the control mutual information to the dependency, ensuring that the dependency is higher when the mutual information is larger; N represents the total number of sampling points; The optimized dependency value range is 0 to 1, where 0 indicates no dependency and 1 indicates complete dependency; the closer the value is to 1, the greater the impact of the system on the overall energy consumption.

5. The method for capturing comprehensive energy data of a low-carbon park according to claim 4, characterized in that: The energy data model is established by constructing a comprehensive energy consumption model based on correlation and dependency, and analyzing the dynamic relationship between different energy types and the overall energy consumption of the park; The overall energy consumption is calculated based on the correlation coefficient and the data of each energy system, expressed as: in, represents the predicted overall energy consumption at time t, m represents the number of energy systems, and w i represents the weight of the i-th system, reflecting its impact on the overall energy consumption; r dynamic,i (t) represents the dynamic correlation coefficient of the ith system at the tth moment, E i (t) represents the energy consumption data of the i-th system at the t-th moment; In dependency analysis, the energy consumption contribution in the model is adjusted according to the dependency of each system on the overall energy consumption. Dependency is used to adjust the influence of each system, which ultimately affects its weight. We perform nonlinear adjustment on the weight in energy consumption prediction, which is expressed as: Among them, w i ′ represents the weight of the ith system after dependency adjustment, D(X i ,Y) represents the dependence of the ith system on the overall energy consumption, w i Represents the initial weight, which is the proportion of each system's impact on the overall energy consumption when not adjusted.

6. The method for capturing comprehensive energy data of a low-carbon park according to claim 5, characterized in that: The establishment of the energy data model also includes introducing a genetic algorithm to optimize the weight w and dependency parameters in the model to make the overall energy consumption prediction more accurate and provide an optimal energy scheduling solution; Randomly generate a set of weights {w i } and the dependency {D(X i ,Y)} candidate solutions, each solution represents a different energy scheduling scheme; Define the fitness function f(w1,w2,…,w m ,D1,D2,…,D m ) is the inverse of the energy consumption prediction error, and the goal is to minimize the prediction error, which can be expressed as: Among them, E true (t) represents the actual energy consumption value, represents the predicted energy consumption value; Select individuals with high fitness as parents and enter the next generation; Perform a crossover operation on the selected parent generation to generate a new candidate solution; The weight w of some individuals that mutate randomly i and the dependency D(X i ,Y), introduce randomness to avoid local optimality; Continue to iterate crossover and mutation until the fitness function converges; The weight w optimized by genetic algorithm i ' and the dependency D(Xi,Y) are used in the final energy consumption prediction model to ensure optimal energy scheduling and energy consumption prediction; Taking into account the correlation, dependency and weight adjustment after genetic algorithm optimization, the final energy consumption prediction is expressed as: Among them, w i ' represents the weight after optimization by genetic algorithm, r dynamic,i (t) represents the dynamic correlation coefficient, E i (t) is the energy consumption data of each system.

7. The method for capturing comprehensive energy data of a low-carbon park according to claim 6, characterized in that: The generating of energy usage optimization suggestions according to the dynamic change relationship includes: According to the correlation and dependence between energy systems and overall energy consumption, a dynamic energy scheduling strategy is formulated to prioritize the scheduling of systems that are highly correlated and highly dependent on overall energy consumption, thereby reducing the use of inefficient energy; During peak load periods, efficient and stable systems are dispatched, and during off-load periods, renewable energy such as photovoltaic systems are used to supplement energy supply, thereby optimizing the overall dispatch of the energy system; Through long-term dependency analysis, we can identify inefficient or unstable systems and make suggestions for maintenance and upgrades. We can regularly maintain inefficient systems to ensure their normal operation. For systems that have been performing inefficiently for a long time, we can make suggestions for hardware upgrades.

8. A low-carbon park comprehensive energy data capture system using the method according to any one of claims 1 to 7, characterized in that: Preprocessing module, collects low-carbon park data and performs preprocessing; An evaluation module that comprehensively evaluates the relevance and dependency of preprocessed data on overall energy consumption; The analysis module builds an energy data model based on the correlation and dependence between data and overall energy consumption, and analyzes the dynamic relationship between different energy types and the overall energy consumption of the park; The suggestion module generates energy usage optimization suggestions based on the dynamically changing relationships.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images