Refrigeration house operation efficiency evaluation method and system combined with energy consumption data analysis

Abstract

The invention discloses a refrigeration house operation efficiency evaluation method and system combined with energy consumption data analysis, and relates to the data processing related field, and the method comprises the steps: collecting refrigeration house operation multi-dimensional data flow; obtaining a historical operation data set of the refrigeration house, performing associated index extraction and dimension reduction analysis, and constructing an energy consumption data index set and a refrigeration house operation efficiency index set; the energy consumption data index set and the refrigeration house operation efficiency index set are adopted to evaluate and classify the refrigeration house historical operation data set, and a refrigeration house energy consumption index data set and a refrigeration house operation efficiency index data set are obtained; and performing mapping evaluation training on the cold storage energy consumption index data set and the cold storage operation efficiency index data set, establishing a cold storage operation efficiency evaluation channel, performing efficiency evaluation on the cold storage operation multi-dimensional data flow, and determining a cold storage operation efficiency evaluation result. The technical problem of insufficient comprehensiveness and accuracy of existing refrigeration house operation efficiency evaluation is solved, and the technical effect of comprehensively and accurately evaluating the refrigeration house operation efficiency is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a cold storage running efficiency evaluation method and system combined with energy consumption data analysis. BACKGROUND

[0002] As a key infrastructure for modern logistics and food storage, the running efficiency of cold storage is directly related to energy consumption cost, goods storage quality and enterprise economic benefit. Efficient and accurate evaluation of cold storage running efficiency is crucial for improving the overall competitiveness and sustainable development of the industry. The current main method to solve the problem of cold storage running efficiency evaluation is to evaluate based on single-dimensional data or simple experience model, such as estimating the energy consumption of cold storage only according to power consumption data, or setting some simple running parameter standards to measure efficiency according to the long-term accumulated experience of staff. The current method cannot fully reflect the complex situation of cold storage running due to only considering single-dimensional data, and the mutual influence between different factors, and single data cannot accurately reflect the overall efficiency. Moreover, simple experience model lacks scientificity and accuracy, and is difficult to adapt to various dynamic changes in the running process of cold storage, resulting in large evaluation result error and unable to provide reliable basis for cold storage optimization running.

[0003] At present, the cold storage running efficiency evaluation in related technology has the technical problems of insufficient comprehensiveness and accuracy. SUMMARY

[0004] The present application provides a cold storage running efficiency evaluation method and system combined with energy consumption data analysis, which collects cold storage running multi-dimensional data stream, obtains cold storage historical running data set, extracts associated indicators and reduces dimension analysis, constructs energy consumption and running efficiency indicator set, evaluates and classifies the historical data set using the above indicator set, obtains cold storage energy consumption and running efficiency indicator data set, maps and evaluates the above two indicator data sets, builds an evaluation channel, and evaluates the cold storage running multi-dimensional data stream accordingly to determine the running efficiency evaluation result. Technical means such as solve the technical problems of insufficient comprehensiveness and accuracy of existing cold storage running efficiency evaluation, and achieve the technical effect of comprehensive and accurate evaluation of cold storage running efficiency.

[0005] The application provides a cold storage operation efficiency evaluation method combined with energy consumption data analysis, including: collecting a cold storage operation multi-dimensional data stream, the cold storage operation multi-dimensional data stream including consumed power data, refrigerant flow data, temperature and humidity data, and pressure data; obtaining a cold storage historical operation data set, performing associated index extraction and dimensionality reduction analysis on the cold storage historical operation data set, constructing an energy consumption data index set and a cold storage operation efficiency index set; using the energy consumption data index set and the cold storage operation efficiency index set to evaluate and classify the cold storage historical operation data set, obtaining a cold storage energy consumption index data set and a cold storage operation efficiency index data set; mapping and evaluating training the cold storage energy consumption index data set and the cold storage operation efficiency index data set, building a cold storage operation efficiency evaluation channel, and based on the cold storage operation efficiency evaluation channel, performing efficiency evaluation on the cold storage operation multi-dimensional data stream to determine a cold storage operation efficiency evaluation result.

[0006] In a possible implementation, the constructing the energy consumption data index set and the cold storage operation efficiency index set performs the following processing: performing data cleaning processing on the cold storage historical operation data set and arranging according to time sequence information to obtain a cold storage sequence operation data set; performing associated index extraction on the cold storage sequence operation data set according to cold storage operation business logic to obtain an energy consumption associated index set and an operation efficiency associated index set; respectively performing cluster analysis on the energy consumption associated index set and the operation efficiency associated index set to obtain an energy consumption index cluster set and an operation efficiency index cluster set; performing index dimensionality reduction analysis based on the energy consumption index cluster set and the operation efficiency index cluster set to obtain the energy consumption data index set and the cold storage operation efficiency index set.

[0007] In a possible implementation, the obtaining the energy consumption index cluster set and the operation efficiency index cluster set performs the following processing: respectively performing associated level division on the energy consumption associated index set and the operation efficiency associated index set to obtain energy consumption index level information and operation efficiency index level information; performing cascade analysis on the energy consumption associated index set and the operation efficiency associated index set according to the energy consumption index level information and the operation efficiency index level information to obtain an energy consumption index cascade tree and an operation efficiency index cascade tree; determining a first-level energy consumption index number and a first-level operation efficiency index number according to the energy consumption index cascade tree and the operation efficiency index cascade tree; using the first-level energy consumption index number and the first-level operation efficiency index number to respectively perform K-means cluster analysis on the energy consumption associated index set and the operation efficiency associated index set to obtain the energy consumption index cluster set and the operation efficiency index cluster set.

[0008] In a possible implementation, the obtaining of the energy consumption data indicator set and the cold storage operation efficiency indicator set comprises: performing decision importance analysis on each indicator cluster in the energy consumption indicator cluster set and the operation efficiency indicator cluster set to obtain an energy consumption indicator cluster decision factor set and an operation efficiency indicator cluster decision factor set; determining an energy consumption indicator cluster dimension reduction coefficient set and an operation efficiency indicator cluster dimension reduction coefficient set according to the energy consumption indicator cluster decision factor set and the operation efficiency indicator cluster decision factor set; and performing indicator dimension reduction processing on the energy consumption indicator cluster set and the operation efficiency indicator cluster set based on the energy consumption indicator cluster dimension reduction coefficient set and the operation efficiency indicator cluster dimension reduction coefficient set to determine the energy consumption data indicator set and the cold storage operation efficiency indicator set.

[0009] In a possible implementation, the obtaining of the energy consumption data indicator set and the cold storage operation efficiency indicator set comprises: obtaining an energy consumption indicator-cluster center distance set and an operation efficiency indicator-cluster center distance set according to the energy consumption indicator cluster set and the operation efficiency indicator cluster set; performing dimension reduction distance analysis on the energy consumption indicator-cluster center distance set and the operation efficiency indicator-cluster center distance set based on the energy consumption indicator cluster dimension reduction coefficient set and the operation efficiency indicator cluster dimension reduction coefficient set to determine an energy consumption indicator cluster dimension reduction distance set and an operation efficiency indicator cluster dimension reduction distance set; and performing indicator dimension reduction processing on the energy consumption indicator cluster set and the operation efficiency indicator cluster set by using the energy consumption indicator cluster dimension reduction distance set and the operation efficiency indicator cluster dimension reduction distance set to determine the energy consumption data indicator set and the cold storage operation efficiency indicator set.

[0010] In a possible implementation, the building of the cold storage operation efficiency evaluation channel comprises: performing association mapping of each efficiency indicator in the cold storage operation efficiency indicator data set with the cold storage energy consumption indicator data set to obtain an operation efficiency indicator-association energy consumption indicator data set; performing efficiency evaluation training on the operation efficiency indicator-association energy consumption indicator data set respectively to generate an operation efficiency indicator evaluation branch channel set; and performing series fusion on the operation efficiency indicator evaluation branch channel set to obtain the cold storage operation efficiency evaluation channel.

[0011] In a possible implementation, the generating of the operation efficiency indicator evaluation branch channel set comprises: selecting a branch indicator channel network structure according to characteristic information of the operation efficiency indicator-association energy consumption indicator data set; and performing efficiency evaluation training on the operation efficiency indicator-association energy consumption indicator data set respectively by using the branch indicator channel network structure to generate the operation efficiency indicator evaluation branch channel set.

[0012] In a possible implementation, the obtaining of the cold storage operation efficiency evaluation channel comprises: performing fusion coefficient evaluation on the set of operation efficiency index evaluation branch channels to determine a set of index branch channel fusion coefficients; and performing serial fusion on the set of operation efficiency index evaluation branch channels based on the set of index branch channel fusion coefficients to obtain the cold storage operation efficiency evaluation channel.

[0013] In a possible implementation, the determining of the set of index branch channel fusion coefficients comprises: performing entropy weight calculation on the set of operation efficiency index evaluation branch channels to determine a branch channel weight distribution coefficient; and determining the set of index branch channel fusion coefficients based on the branch channel weight distribution coefficient and output precision of the set of operation efficiency index evaluation branch channels.

[0014] The application further provides a cold storage operation efficiency evaluation system combined with energy consumption data analysis, comprising: a cold storage operation multi-dimensional data stream acquisition module, configured to acquire a cold storage operation multi-dimensional data stream, wherein the cold storage operation multi-dimensional data stream comprises consumed power data, refrigerant flow data, temperature and humidity data, and pressure data; an index set construction module, configured to acquire a cold storage historical operation data set, perform associated index extraction and dimension reduction analysis on the cold storage historical operation data set, and construct an energy consumption data index set and a cold storage operation efficiency index set; an index data set acquisition module, configured to perform evaluation classification on the cold storage historical operation data set by using the energy consumption data index set and the cold storage operation efficiency index set, and obtain a cold storage energy consumption index data set and a cold storage operation efficiency index data set; and an efficiency evaluation module, configured to perform mapping evaluation training on the cold storage energy consumption index data set and the cold storage operation efficiency index data set, build a cold storage operation efficiency evaluation channel, perform efficiency evaluation on the cold storage operation multi-dimensional data stream based on the cold storage operation efficiency evaluation channel, and determine a cold storage operation efficiency evaluation result.

[0015] The application provides a cold storage operation efficiency evaluation method and system combined with energy consumption data analysis. First, a cold storage operation multi-dimensional data stream is acquired, wherein the cold storage operation multi-dimensional data stream comprises consumed power data, refrigerant flow data, temperature and humidity data, and pressure data. Then, a cold storage historical operation data set is acquired, associated index extraction and dimension reduction analysis are performed on the cold storage historical operation data set, an energy consumption data index set and a cold storage operation efficiency index set are constructed. Next, evaluation classification is performed on the cold storage historical operation data set by using the energy consumption data index set and the cold storage operation efficiency index set, and a cold storage energy consumption index data set and a cold storage operation efficiency index data set are obtained. Finally, mapping evaluation training is performed on the cold storage energy consumption index data set and the cold storage operation efficiency index data set, a cold storage operation efficiency evaluation channel is built, efficiency evaluation is performed on the cold storage operation multi-dimensional data stream based on the cold storage operation efficiency evaluation channel, and a cold storage operation efficiency evaluation result is determined. The cold storage operation efficiency can be comprehensively and accurately evaluated. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a flowchart illustrating a method for evaluating the operating efficiency of a cold storage facility by combining energy consumption data analysis, as provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of a cold storage operation efficiency evaluation system that combines energy consumption data analysis, provided as an embodiment of this application.

[0019] Figure labeling: 10 for multi-dimensional data stream acquisition module for cold storage operation, 20 for indicator set construction module, 30 for indicator dataset acquisition module, and 40 for efficiency evaluation module. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to only distinguishes similar objects, and does not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiments of the present application provide a cold storage running efficiency evaluation method combined with energy consumption data analysis, as shown in Figure 1 The method comprises the following steps: Step S100, collecting cold storage running multi-dimensional data stream, wherein the cold storage running multi-dimensional data stream comprises consumed power data, refrigerant flow data, temperature and humidity data, and pressure data.

[0024] Specifically, different types of data are collected by installing various types of sensors in the cold storage. For the consumed power data, a smart power meter is used, which can accurately measure the power consumption of the cold storage equipment in real time and output the data in the form of digital signals; the refrigerant flow data is obtained by installing a flow sensor on the refrigerant pipeline, which can sense the flow speed and flow size of the refrigerant based on specific flow measurement principles such as vortex street, electromagnetic, etc.; the temperature and humidity data are collected using a temperature and humidity sensor, which can measure the temperature and humidity inside the cold storage at the same time and convert the analog signal into a digital signal; the pressure data are collected by a pressure sensor, which is installed at key pressure points of the cold storage refrigeration system, such as the outlet of the compressor and the inlet of the condenser, to measure the pressure values of the refrigerant at different positions. These sensors transmit the collected data to the data acquisition terminal through wired or wireless means.

[0025] For example, in a large cold storage, multiple temperature and humidity sensors are evenly distributed in each area of the cold storage to comprehensively obtain the temperature and humidity conditions at different positions inside the cold storage. A smart power meter is installed at the main distribution box of the cold storage to record the power consumption of the entire cold storage. A flow sensor is installed on the main refrigerant pipeline to monitor the flow changes of the refrigerant in real time. Pressure sensors are installed at the outlet of the compressor and the inlet of the condenser, respectively.

[0026] In step S200, a cold storage historical operation data set is obtained, and the cold storage historical operation data set is subjected to correlation index extraction and dimension reduction analysis to construct an energy consumption data index set and a cold storage operation efficiency index set.

[0027] Specifically, a cold storage historical operation data set is obtained from a historical database of the cold storage, and the data is stored in a relational database or a time series database. The correlation index extraction adopts a data mining algorithm, such as an association rule mining algorithm, to find out key indexes related to energy consumption and operation efficiency by analyzing the correlation between different parameters in the historical data. For example, the correlation between power consumption and refrigerant flow, temperature and humidity changes is analyzed to determine which factors have a significant impact on energy consumption. The dimension reduction analysis can use dimension reduction methods such as principal component analysis (PCA) algorithm or linear discriminant analysis (LDA) algorithm. The PCA algorithm projects high-dimensional data into a low-dimensional space through linear transformation of the data, retains the main features of the data, reduces the data dimension, and reduces the computational complexity. The LDA algorithm finds a projection direction under the consideration of data class information, so that the projections of different classes of data in the direction are as far apart as possible, and the data of the same class is as close as possible. After correlation index extraction and dimension reduction analysis, the energy consumption data index set and the cold storage operation efficiency index set are constructed according to business requirements and data analysis results. The energy consumption data index set can include unit time energy consumption, unit cargo storage energy consumption, etc.; the cold storage operation efficiency index set can include refrigeration efficiency, temperature stability, etc.

[0028] In one possible implementation, the step S200 of constructing the energy consumption data index set and the cold storage operation efficiency index set further includes a step S210 of performing data cleaning processing on the cold storage historical operation data set and arranging the data according to time sequence information to obtain a cold storage sequence operation data set. Specifically, a plurality of data cleaning rules and algorithms are used to process problem data in the cold storage historical operation data set. For missing values, if the missing proportion is small, methods such as mean filling, median filling or filling with adjacent data before and after can be used; if the missing proportion is large, the data record can be deleted. For abnormal values, a reasonable threshold range is set for identification, for example, for temperature data, upper and lower limits are set according to the normal operation temperature range of the cold storage, and data exceeding the range is considered as abnormal value, which can be corrected by smoothing or directly removed. For duplicate data, completely duplicate records are identified and deleted by comparing the field values of the data records.

[0029] According to the time stamp information recorded in the data, a sorting algorithm is used to arrange the cleaned data in chronological order to ensure the continuity and order of the data in the time dimension, thereby obtaining the cold storage sequence operation data set.

[0030] Step S220, according to the cold storage operation business logic, the cold storage sequence operation data set is associated with index extraction, and the energy consumption associated index set and the operation efficiency associated index set are obtained. Specifically, the operation process and characteristics of the cold storage are analyzed, and the key factors and business rules affecting energy consumption and operation efficiency are determined. For example, the energy consumption of the cold storage is closely related to the running time of the refrigeration equipment, the flow of the refrigerant, the temperature difference between the inside and outside of the cold storage, etc.; the operation efficiency is related to the refrigeration capacity of the refrigeration system, the stability of temperature control, the storage mode of goods, etc. Based on the above business logic, the association rule mining algorithm in data mining is used to find out the indexes related to energy consumption and operation efficiency from the cold storage sequence operation data set. For energy consumption associated indexes, it can include unit time refrigeration equipment running power, refrigerant consumption per unit of goods storage, etc.; for operation efficiency associated indexes, it can include temperature fluctuation range of the refrigeration system within a certain time, refrigerant circulation efficiency, etc.

[0031] For example, by analyzing the cold storage operation business logic, it is found that when the number of cold storage door openings increases, the temperature inside the cold storage will rise, resulting in the extension of the running time of the refrigeration equipment and the increase of energy consumption. Therefore, the number of cold storage door openings is taken as an index related to energy consumption. At the same time, it is found that the heat dissipation effect of the condenser in the refrigeration system will affect the circulation efficiency of the refrigerant, and then affect the operation efficiency of the cold storage, so the heat dissipation temperature difference of the condenser is taken as one of the operation efficiency associated indexes.

[0032] Step S230, respectively, the energy consumption associated index set and the operation efficiency associated index set are subjected to cluster analysis, and the energy consumption index cluster set and the operation efficiency index cluster set are obtained. Specifically, clustering algorithms such as K-means clustering algorithm, hierarchical clustering algorithm, etc. are used. The K-means clustering algorithm divides the data points into K clusters through iteration, so that each data point has the minimum distance to the center of the cluster it belongs to; the hierarchical clustering algorithm constructs a hierarchical structure by continuously merging or splitting clusters, and finally obtains the cluster division at different levels.

[0033] For the energy consumption associated index set, each energy consumption associated index is taken as a data point, and cluster analysis is performed according to the selected clustering algorithm. For example, when using the K-means clustering algorithm, first, K initial cluster centers are randomly selected, then each data point is assigned to the cluster where the nearest cluster center is located, then the center of each cluster is recalculated, and the above assignment and calculation process is repeated until the cluster center no longer changes or the preset iteration number is reached, and the energy consumption index cluster set is obtained, each cluster respectively represents different energy consumption characteristic modes, such as high energy consumption mode, medium energy consumption mode and low energy consumption mode. Similarly, the operation efficiency associated index set is subjected to cluster analysis, and the operation efficiency index cluster set is obtained.

[0034] Step S240: Based on the energy consumption index cluster set and the operational efficiency index cluster set, perform dimensionality reduction analysis to obtain the energy consumption data index set and the cold storage operational efficiency index set. Specifically, taking the PCA algorithm as an example, for the energy consumption index cluster set, first calculate the covariance matrix between each energy consumption index, then find the eigenvalues ​​and eigenvectors of the covariance matrix, and select the first few eigenvectors as projection directions according to the eigenvalues ​​in descending order. Project the original energy consumption index data onto these directions to obtain the dimensionality-reduced energy consumption data index set. Similarly, perform dimensionality reduction analysis on the operational efficiency index cluster set to obtain the cold storage operational efficiency index set.

[0035] For example, PCA algorithm is used to perform dimensionality reduction analysis on a set of energy consumption index clusters. After calculating the covariance matrix, it is found that the first two eigenvalues ​​account for more than 90% of the total eigenvalues, indicating that these two eigenvectors can retain most of the information of the original data. Therefore, these two eigenvectors are selected as projection directions, and the original multiple energy consumption index data are projected onto these two directions to obtain two comprehensive energy consumption indexes, forming an energy consumption data index set.

[0036] This approach effectively removes missing, outlier, and duplicate data from the historical cold storage operation dataset through data cleaning, ensuring the accuracy and integrity of the data. Arranging the data according to time sequence information gives it continuity and orderliness over time, providing a reliable foundation for data analysis and processing.

[0037] In one possible implementation, obtaining the energy consumption index cluster set and the operating efficiency index cluster set, step S230 further includes step S231, which involves hierarchically dividing the energy consumption index cluster set and the operating efficiency index cluster set to obtain hierarchical information of energy consumption indicators and operating efficiency indicators, respectively. Specifically, by analyzing various aspects of energy consumption and operating efficiency during cold storage operation, and combining industry expert experience and the characteristics of actual operating data, the logical relationships and hierarchical structure between indicators are identified. For example, for energy consumption related indicators, starting from the source and influencing factors of energy consumption, they can be divided into equipment operation energy consumption related indicators, environmental factor-related energy consumption indicators, and cargo storage-related energy consumption indicators. A hierarchical analysis method is adopted to divide the related indicators according to the principle of moving from macro to micro and from general to specific.

[0038] Taking operational efficiency-related indicators as an example, the first layer can be the overall operational efficiency indicator; the second layer can be divided into refrigeration system efficiency indicators, electrical system efficiency indicators, etc.; and the third layer can be further subdivided, such as the refrigeration system efficiency indicator being subdivided into evaporator efficiency indicators, condenser efficiency indicators, etc. In this way, hierarchical information on energy consumption indicators and operational efficiency indicators can be constructed.

[0039] Step S232: Perform cascade analysis on the energy consumption-related indicator set and the operation efficiency-related indicator set according to the energy consumption indicator hierarchy information and the operation efficiency indicator hierarchy information to obtain the energy consumption indicator cascade tree and the operation efficiency indicator cascade tree. Specifically, based on the hierarchical division results, a tree data structure is used to represent the relationships between indicators. The root node represents the overall concept, such as energy consumption or operation efficiency, and each child node represents the next level indicator, forming a cascade tree through the parent-child node relationship. For example, the root node of the energy consumption indicator cascade tree is "energy consumption," and its child nodes are "equipment operation energy consumption," "environmental impact energy consumption," and "cargo storage energy consumption," with each child node having its own child nodes, and so on. Similarly, the operation efficiency indicator cascade tree is constructed.

[0040] In constructing the cascading tree, the relationships between each indicator and its superior and subordinate indicators are analyzed. Superior indicators provide a general overview and guidance for subordinate indicators, while subordinate indicators are the specific manifestations and subdivisions of superior indicators. This cascading analysis reveals the hierarchical structure and logical connections between indicators.

[0041] Step S233: Based on the energy consumption index cascade tree and the operational efficiency index cascade tree, determine the number of first-level energy consumption indicators and the number of first-level operational efficiency indicators. Specifically, first-level indicators are child nodes of the root node, possessing generality and specificity, and representing the main characteristics and key factors of the field. In the energy consumption index cascade tree, first-level indicators include the main sources and influencing factors of energy consumption; in the operational efficiency index cascade tree, first-level indicators include the main aspects and key links of operational efficiency. The number of first-level indicators is determined through analysis of the cascade trees.

[0042] For example, in the cascaded tree of operational efficiency indicators, the first-level indicators include three main aspects: cooling effect, equipment operation, and cargo-related aspects. These three aspects are relatively independent and representative. Therefore, the number of first-level operational efficiency indicators is determined to be 3, corresponding to the cooling effect indicator, equipment operation indicator, and cargo-related indicator, respectively. Similarly, the number of first-level energy consumption indicators is determined.

[0043] Step S234: K-means clustering analysis is performed on the energy consumption-related indicator set and the operation efficiency-related indicator set using the number of first-level energy consumption indicators and the number of first-level operation efficiency indicators, respectively, to obtain the energy consumption indicator cluster set and the operation efficiency indicator cluster set. Specifically, the K-means clustering algorithm is a distance-based clustering algorithm whose goal is to divide data points into K clusters such that the distance from each data point to the center of its cluster is minimized. In this application, the value of K is taken as the number of first-level energy consumption indicators and the number of first-level operation efficiency indicators, respectively.

[0044] For the set of energy consumption-related indicators, each energy consumption indicator is treated as a data point. Based on the selected K value, i.e., the number of first-level energy consumption indicators, K-means clustering analysis is performed. First, K initial cluster centers are randomly selected. Then, each data point is assigned to the cluster containing the nearest cluster center. Next, the center of each cluster is recalculated, and the above assignment and calculation process is repeated until the cluster centers no longer change or the preset number of iterations is reached. Similarly, clustering analysis is performed on the set of operational efficiency-related indicators to obtain a set of operational efficiency indicator clusters.

[0045] This implementation uses the number of first-level indicators as the number of clusters, combining business knowledge and data characteristics to make the clustering results more consistent with reality. First-level indicators are general and specific, representing the main characteristics and key factors of the field. Clustering analysis based on these indicators can improve the rationality and accuracy of clustering and avoid the problems caused by blindly selecting the number of clusters.

[0046] In one possible implementation, obtaining the energy consumption data index set and the cold storage operation efficiency index set, step S240 further includes step S241, which involves performing a decision importance analysis on each index cluster in the energy consumption index cluster set and the operation efficiency index cluster set to obtain a decision factor set for energy consumption index clusters and a decision factor set for operation efficiency index clusters. Specifically, experts in cold storage operation management, energy management, and data analysis are invited to assess the importance of each energy consumption index cluster and operation efficiency index cluster based on their professional knowledge and practical experience. Experts can score or rank the index clusters based on their impact on cold storage energy consumption and operation efficiency, their criticality, and their controllability. Simultaneously, historical data is used to evaluate the importance of each index cluster through statistical analysis methods. The correlation coefficient between each index cluster and cold storage energy consumption or operation efficiency can be calculated; the higher the correlation coefficient, the greater the impact of the index cluster on energy consumption or operation efficiency, and the higher its importance. By combining expert evaluation and data analysis methods, and integrating expert opinions and data analysis results, the decision factors for each index cluster are determined. For example, the expert evaluation results and data analysis results can be assigned certain weights, and then a weighted average can be performed to obtain the final decision factor.

[0047] Step S242: Based on the decision factor set of the energy consumption index cluster and the decision factor set of the operation efficiency index cluster, determine the dimensionality reduction coefficient set of the energy consumption index cluster and the dimensionality reduction coefficient set of the operation efficiency index cluster. Specifically, the larger the decision factor, the more important the index cluster, the more indicators need to be retained, and the smaller the corresponding dimensionality reduction coefficient. A mapping relationship between the decision factor and the dimensionality reduction coefficient can be set, such as using a linear mapping or a nonlinear mapping. A linear mapping can be expressed as: dimensionality reduction coefficient = a − b × decision factor, where a and b are constants determined according to the actual situation to ensure that the dimensionality reduction coefficient is within a reasonable range. Nonlinear mapping can be designed according to specific needs, such as using exponential functions, logarithmic functions, etc. Different parameter combinations can be tried through experiments or simulations to observe the impact of the dimensionality-reduced index set on the evaluation of cold storage energy consumption and operation efficiency, select the optimal parameter combination, and adjust and optimize the parameters in the mapping relationship.

[0048] Step S243: Based on the set of dimensionality reduction coefficients for the energy consumption index clusters and the set of dimensionality reduction coefficients for the operating efficiency index clusters, perform dimensionality reduction processing on the energy consumption index cluster set and the operating efficiency index cluster set to determine the energy consumption data index set and the cold storage operating efficiency index set. Specifically, for each index cluster, the distance between each index and the cluster center is calculated. Distance calculation methods include Euclidean distance, Manhattan distance, etc. Dimensionality reduction is performed according to the dimensionality reduction coefficient and the distance between the index and the cluster center. The indexes are sorted in descending order of distance. For each index cluster, the number of indicators to be retained is determined based on its dimensionality reduction coefficient. For example, if the dimensionality reduction coefficient of an index cluster is 5.8, it can be rounded down to 5, that is, the 5 indicators closest to the cluster center are retained, and the remaining indicators are subjected to dimensionality reduction processing, such as deletion or merging. After performing the above dimensionality reduction operation on all energy consumption index clusters and operating efficiency index clusters, the retained indicators are combined to form the energy consumption data index set and the cold storage operating efficiency index set, respectively.

[0049] This approach, through decision importance analysis, accurately identifies key indicator clusters that significantly impact cold storage energy consumption and operational efficiency, and prioritizes retaining important indicators from these clusters during dimensionality reduction. This ensures that the reduced indicator set still reflects the core characteristics of cold storage operation, improving the accuracy and reliability of the assessment.

[0050] In one possible implementation, the step S243 of determining the energy consumption data index set and the cold storage operation efficiency index set further includes step S2431, which involves obtaining the energy consumption index-cluster center distance set and the operation efficiency index-cluster center distance set based on the energy consumption index cluster set and the operation efficiency index cluster set. Specifically, for each index cluster in the energy consumption index cluster set, its cluster center is calculated. Similarly, the cluster center of each index cluster in the operation efficiency index cluster set is calculated. The distance between each index and its cluster center is calculated using a distance metric method. The distances between all indicators in the energy consumption index cluster set and their respective cluster centers are calculated sequentially to obtain the energy consumption index-cluster center distance set; similarly, the operation efficiency index-cluster center distance set is calculated.

[0051] Step S2432: Based on the set of dimensionality reduction coefficients for the energy consumption index clusters and the set of dimensionality reduction coefficients for the operation efficiency index clusters, perform dimensionality reduction distance analysis on the set of distances between the energy consumption indexes and the cluster centers, and determine the set of dimensionality reduction distances for the energy consumption index clusters and the operation efficiency index clusters. Specifically, based on the set of dimensionality reduction coefficients for the energy consumption index clusters and the operation efficiency index clusters, determine the number of indicators to be retained in each index cluster. The smaller the dimensionality reduction coefficient, the more important the index cluster is, and the more indicators need to be retained. A threshold or proportion related to the dimensionality reduction coefficient can be set to determine the number of indicators to be retained. For each index cluster, sort the indicators in descending order of their distance from the cluster center. Based on the determined number of indicators to be retained, select the indicators that are closer to the cluster center and retain their corresponding distances to form a subset of the dimensionality reduction distances for that index cluster. Combine the subsets of the dimensionality reduction distances for all energy consumption index clusters and operation efficiency index clusters to obtain the set of dimensionality reduction distances for the energy consumption index clusters and the set of dimensionality reduction distances for the operation efficiency index clusters, respectively.

[0052] Step S2433: Dimensionality reduction processing is performed on the energy consumption index cluster set and the operation efficiency index cluster set using the dimensionality reduction distance set of the energy consumption index cluster and the operation efficiency index cluster set, to determine the energy consumption data index set and the cold storage operation efficiency index set. Specifically, based on the dimensionality reduction distance set of the energy consumption index cluster and the operation efficiency index cluster, the indicators to be retained in each index cluster are determined. For each energy consumption index cluster, a subset of the dimensionality reduction distances corresponding to each index cluster is found in the dimensionality reduction distance set; the indicators corresponding to the distances in these subsets are the indicators to be retained. Similarly, the indicators to be retained in the operation efficiency index cluster are determined. The indicators retained from all energy consumption index clusters are combined to form the energy consumption data index set; the indicators retained from all operation efficiency index clusters are combined to form the cold storage operation efficiency index set.

[0053] This approach calculates the distance between the indicators and the cluster center, and performs dimensionality reduction distance analysis based on the dimensionality reduction coefficient. This allows for a more accurate determination of which indicators to retain and which to discard. Indicators closer to the cluster center are more representative of the characteristics of that indicator cluster. Retaining these indicators preserves the information in the original data to the greatest extent possible, while effectively reducing the number of indicators and improving data processing efficiency.

[0054] Step S300: The energy consumption data index set and the cold storage operation efficiency index set are used to evaluate and classify the historical operation dataset of the cold storage to obtain the cold storage energy consumption index dataset and the cold storage operation efficiency index dataset.

[0055] Specifically, classification algorithms are used to evaluate and classify the historical operation dataset of cold storage facilities. These algorithms can include decision tree algorithms, support vector machines (SVM), etc. Taking the decision tree algorithm as an example, based on the feature values ​​of the energy consumption data index set and the cold storage operation efficiency index set, the historical operation dataset is recursively divided into smaller subsets to construct a tree-structured classification model. Each internal node represents a test on a feature, each branch represents a test output, and each leaf node represents a category. By training the decision tree model, it can classify the historical operation data of cold storage facilities into different energy consumption levels and operation efficiency levels based on the input energy consumption and operation efficiency index data. The classified data is then organized according to energy consumption and operation efficiency indices to obtain the cold storage energy consumption index dataset and the cold storage operation efficiency index dataset.

[0056] Step S400: Map and evaluate the cold storage energy consumption index dataset and the cold storage operation efficiency index dataset to build a cold storage operation efficiency evaluation channel, and evaluate the efficiency of the cold storage operation multidimensional data stream based on the cold storage operation efficiency evaluation channel to determine the cold storage operation efficiency evaluation result.

[0057] Specifically, regression algorithms from machine learning, such as linear regression or neural network algorithms, are used for mapping evaluation training. Taking neural network algorithms as an example, a multilayer perceptron (MLP) neural network model is constructed, and the cold storage energy consumption index dataset and the cold storage operation efficiency index dataset are used as input and output samples to train the neural network. By adjusting the weights and biases of the neural network, the model can learn the mapping relationship between energy consumption indexes and operation efficiency indexes. After training, a cold storage operation efficiency evaluation channel is built. This channel, with the neural network model as its core, inputs the preprocessed multidimensional data stream of real-time collected cold storage operation data into the neural network model, and the model outputs the cold storage operation efficiency evaluation results.

[0058] For example, an MLP neural network model is constructed, comprising an input layer, hidden layers, and an output layer. The number of nodes in the input layer is determined based on the feature dimensions of the cold storage energy consumption index dataset and the cold storage operation efficiency index dataset. The number of nodes in the hidden layer is determined through experiments and optimization. The output layer has 1 node, representing the cold storage operation efficiency evaluation value. Historical data is used to train the neural network, and the weights and biases are adjusted through multiple iterations to minimize the model's prediction error. In practical applications, the real-time collected multi-dimensional cold storage operation data stream is preprocessed and input into the trained neural network model. The model outputs a value between 0 and 1 as the cold storage operation efficiency evaluation result; the closer the value is to 1, the higher the operation efficiency.

[0059] In one possible implementation, the establishment of the cold storage operation efficiency evaluation channel, step S400 further includes step S410, which involves associating each efficiency indicator in the cold storage operation efficiency indicator dataset with the cold storage energy consumption indicator dataset to obtain an operation efficiency indicator-associated energy consumption indicator dataset. Specifically, data feature analysis is performed on the cold storage operation efficiency indicator dataset and the cold storage energy consumption indicator dataset to determine the meaning, dimensions, value range, physical meaning, and interrelationships of each efficiency indicator and energy consumption indicator during cold storage operation. Based on the data feature analysis results, association rules between each efficiency indicator and energy consumption indicator are determined. Correlation analysis methods, such as Pearson correlation coefficient and Spearman rank correlation coefficient, can be used to calculate the correlation between efficiency indicators and energy consumption indicators. For indicator pairs with high correlation, a clear association mapping relationship is established. According to the determined association rules, each efficiency indicator in the cold storage operation efficiency indicator dataset is combined with its corresponding associated energy consumption indicator to form an operation efficiency indicator-associated energy consumption indicator dataset.

[0060] Step S420: Efficiency evaluation training is performed on the operational efficiency index-related energy consumption index datasets to generate an operational efficiency index evaluation branch channel set. Specifically, for each operational efficiency index-related energy consumption index dataset, an evaluation model is selected for training, including linear regression models, neural network models, support vector machine models, etc. The operational efficiency index-related energy consumption index dataset is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used to adjust the model's hyperparameters and prevent overfitting, and the test set is used to evaluate the model's final performance. A typical ratio is 6:2:2 or 7:1:2. The selected model is trained using the training set, and the model's parameters are continuously adjusted to accurately fit the relationship between the operational efficiency index and the related energy consumption index. During training, the model is evaluated using the validation set, and the model's hyperparameters, such as the number of neural network layers, the number of neurons, and the learning rate, are adjusted based on the evaluation results to improve the model's generalization ability. After multiple training and optimizations, a high-performance evaluation model is obtained. An operational efficiency index evaluation branch channel is generated for each trained evaluation model. Each branch channel is an independent evaluation unit used to evaluate specific operational efficiency indicators and output corresponding evaluation results.

[0061] Step S430: The set of operational efficiency indicator evaluation branch channels is fused in series to obtain a cold storage operational efficiency evaluation channel. Specifically, based on the importance of each operational efficiency indicator evaluation branch channel and its comprehensive impact on cold storage operational efficiency, a corresponding weight is assigned to each branch channel. Weights can be determined using methods such as expert scoring or analytic hierarchy process (AHP). The various operational efficiency indicator evaluation branch channels are then fused in series in a certain order. A weighted summation method can be used, multiplying the evaluation results of each branch channel by its corresponding weight and then summing them to obtain a comprehensive evaluation value for cold storage operational efficiency. Alternatively, other fusion strategies can be used, such as a fusion method based on fuzzy comprehensive evaluation, considering the uncertainty and fuzziness of the evaluation results of each branch channel. Based on the selected series fusion strategy, the set of operational efficiency indicator evaluation branch channels is integrated to construct the cold storage operational efficiency evaluation channel.

[0062] This approach, by training and evaluating the branch channels separately, fully considers the complex relationships between various indicators during cold storage operation, and evaluates cold storage operational efficiency from multiple perspectives. By cascading and fusing the results of each branch channel, a more comprehensive and accurate evaluation value of cold storage operational efficiency can be obtained, avoiding the one-sidedness of evaluation based on a single indicator.

[0063] In one possible implementation, step S420, which generates the operational efficiency index evaluation branch channel set, further includes step S421, selecting the branch index channel network structure based on the characteristic information of the operational efficiency index-related energy consumption index dataset. Specifically, firstly, data characteristic analysis is performed, including exploring the data distribution of the operational efficiency index-related energy consumption index dataset by drawing statistical charts such as histograms and box plots to observe the distribution pattern of the data and determine whether it is a normal distribution, a skewed distribution, or other complex distribution; using correlation analysis methods, such as Pearson correlation coefficient and Spearman rank correlation coefficient, the correlation between operational efficiency index and related energy consumption index is calculated to determine whether they are linearly correlated, nonlinearly correlated, or have no obvious correlation; the dimensionality and sample size of the data set are analyzed to determine the complexity of the data and the amount of data that can be used to train the model.

[0064] Network structure selection is based on data characteristic analysis results. If there is a clear linear relationship between operating efficiency indicators and associated energy consumption indicators, a simple linear network structure, such as a single-layer perceptron, can be chosen. A single-layer perceptron can model linear relationships through simple linear combinations and activation functions, offering advantages such as computational simplicity and fast training speed. For datasets with complex nonlinear relationships, deep neural network structures can be selected, such as multilayer perceptrons (MLPs), convolutional neural networks (CNNs), or recurrent neural networks (RNNs). Multilayer perceptrons, through nonlinear transformations of multiple hidden layers, can fit complex nonlinear functions; convolutional neural networks can extract local features of data using convolutional kernels, making them suitable for data with local correlations; and recurrent neural networks can handle time-series data, capturing time dependencies within the data. If the dataset contains uncertainty and ambiguity, such as data fluctuations due to sensor errors or changes in the measurement environment, a fuzzy neural network can be chosen. Fuzzy neural networks combine the advantages of fuzzy logic and neural networks, enabling them to handle uncertain and fuzzy information and improve the robustness of the model.

[0065] Step S422: The branch indicator channel network structure is used to train the efficiency evaluation of the operational efficiency indicator-related energy consumption indicator dataset, generating the operational efficiency indicator evaluation branch channel set. Specifically, the model parameters are initialized according to the selected branch indicator channel network structure, using methods such as random initialization, Xavier initialization, and He initialization. A suitable loss function is selected based on the evaluation task objective, such as mean squared error (MSE) or mean absolute error (MAE). A suitable optimization algorithm is selected to update the model parameters to minimize the loss function; optimization algorithms include stochastic gradient descent (SGD), Adam, and RMSprop.

[0066] The runtime efficiency metric-related energy consumption metric dataset is divided into training, validation, and test sets. During training, the model is periodically evaluated using the validation set, and its hyperparameters are adjusted based on the evaluation results. If the model performs poorly on the validation set, measures such as adjusting the network structure, increasing training data, and changing the optimization algorithm can be attempted to improve the model. After multiple iterations of training and adjustments, a high-performing model is obtained and serves as the runtime efficiency metric evaluation branch channel. Similar operations are performed on each runtime efficiency metric-related energy consumption metric dataset to generate multiple runtime efficiency metric evaluation branch channels, forming a runtime efficiency metric evaluation branch channel set.

[0067] This approach, by selecting an appropriate branch indicator channel network structure based on the characteristics of the operational efficiency indicator-correlated energy consumption indicator dataset, can better fit the complex relationships between indicators. Different network structures have different advantages, thereby improving the accuracy of cold storage operational efficiency assessment.

[0068] In one possible implementation, obtaining the cold storage operation efficiency evaluation channel, step S430 further includes step S431, evaluating the fusion coefficients of the operation efficiency index evaluation branch channel set, and determining the fusion coefficient set of index branch channels. Specifically, the fusion coefficients are determined based on the importance of the indicators using the Analytic Hierarchy Process (AHP) or the entropy weight method. Taking AHP as an example, firstly, a hierarchical structure model is constructed. The cold storage operation efficiency evaluation is taken as the target layer, and each operation efficiency index evaluation branch channel is taken as the criterion layer. Then, a judgment matrix is ​​constructed. Through expert scoring, the importance of each branch channel in the criterion layer relative to the target layer is compared pairwise to construct the judgment matrix. Next, a hierarchical single ranking and consistency test are performed. The largest eigenvalue of the judgment matrix and its corresponding eigenvector are calculated. After normalizing the eigenvector, the relative importance weight of each branch channel relative to the target layer is obtained, i.e., the preliminary value of the fusion coefficient. At the same time, the consistency index (CI) and the random consistency index (RI) are calculated, and the consistency ratio (CR=CI / RI) is obtained. When CR < 0.1, the judgment matrix is ​​considered to have satisfactory consistency, and the obtained weight values ​​are reasonable; otherwise, the judgment matrix needs to be adjusted, and the weights recalculated. Finally, after multiple adjustments and verifications, the fusion coefficients of each operational efficiency index evaluation branch channel are determined, forming a set of index branch channel fusion coefficients.

[0069] Step S432: Based on the set of fusion coefficients for the indicator branch channels, the set of operational efficiency indicator evaluation branch channels is fused in series to obtain the cold storage operational efficiency evaluation channel. Specifically, the output of each operational efficiency indicator evaluation branch channel is an evaluation value, which can be a numerical efficiency score or a numerical code corresponding to a categorized efficiency level. The evaluation values ​​of each branch channel are multiplied by their corresponding fusion coefficients and then summed to obtain the final evaluation value of the cold storage operational efficiency.

[0070] This approach integrates information from multiple operational efficiency indicators across various evaluation channels, taking into account several key aspects of cold storage operation. Compared to evaluation based on a single indicator, it provides a more comprehensive reflection of the overall operational efficiency of the cold storage and avoids evaluation biases caused by focusing on only one indicator.

[0071] In one possible implementation, the step S431 of determining the set of fusion coefficients for the indicator branch channels further includes step S4311, which involves calculating entropy weights on the set of branch channels for evaluating the operational efficiency indicator to determine the weight allocation coefficients for the branch channels. Specifically, the entropy weight method is an objective weighting method based on information entropy. In information theory, entropy is a measure of uncertainty. For an evaluation system, the smaller the information entropy of an indicator, the greater the amount of information it provides, and the greater its role in the evaluation, thus its weight should be higher; conversely, the greater the information entropy, the less information the indicator provides, and the lower its weight.

[0072] Assume there are m branch channels in the operational efficiency evaluation branch channel set, which are sequentially labeled as Channel 1, Channel 2, and so on up to Channel m. Simultaneously, data from n evaluation samples are collected. These samples can come from operational data of cold storage facilities at different time points or from operational data of cold storage facilities in different regions. The evaluation results corresponding to each branch channel are organized into a data matrix. Specifically, for each sample (a total of n samples), it will have an evaluation value under each of the m branch channels. The evaluation value of the i-th sample under the j-th branch channel is recorded and arranged in the order of sample first, branch channel last, forming an n x m data matrix. In this matrix, each row represents the evaluation status of a sample under each branch channel, and each column represents the evaluation status of all samples under a given branch channel.

[0073] Because different branch channels may evaluate different content and methods, their evaluation values ​​will have different dimensions and orders of magnitude. For example, some branch channels evaluate indicators presented as percentages, while others evaluate indicators presented as specific numerical values, which may differ significantly. To eliminate the impact of these differences in dimensions and orders of magnitude, data standardization is necessary. For positive indicators, i.e., indicators where a larger evaluation value indicates higher cold storage operating efficiency, the following standardization method is used: First, find the maximum and minimum values ​​of all sample evaluation values ​​under the j-th branch channel. Then, subtract the minimum value from the evaluation value of the i-th sample under the j-th branch channel, and divide by the difference between the maximum and minimum values. This calculation converts the original evaluation value into a value between 0 and 1. For negative indicators, i.e., indicators where a smaller evaluation value indicates higher cold storage operating efficiency, the standardization method is different. Similarly, first find the maximum and minimum values ​​of all sample evaluation values ​​under the j-th branch channel. Then, subtract the evaluation value of the i-th sample under the j-th branch channel from the maximum value, and divide by the difference between the maximum and minimum values. This process also normalizes the original evaluation values ​​to the range of 0 to 1. After this standardization process, a new data matrix is ​​obtained. In this new matrix, each element is within the range of 0 to 1, eliminating the influence of dimensions and orders of magnitude.

[0074] To calculate information entropy, the standardized data needs to be transformed into a probability distribution. Specifically, for the j-th branch channel, calculate the proportion of the evaluation value of the i-th sample in that channel to the sum of the evaluation values ​​of all samples in that channel. In other words, divide the standardized evaluation value of the i-th sample in the j-th branch channel by the sum of the standardized evaluation values ​​of all samples in the j-th branch channel; the result is the proportion of that sample in that branch channel. Through this calculation, the sample evaluation values ​​in each branch channel constitute a probability distribution.

[0075] Information entropy measures the degree of uncertainty in the j-th branch channel. It is calculated based on the weight of each sample in that branch channel, obtained from previous calculations. Specifically, the weight of each sample in the j-th branch channel is multiplied by the natural logarithm of that weight. Then, this calculation is summed for all samples, and finally multiplied by an adjustment factor k. This adjustment factor k is determined by the sample size n, and its value is 1 divided by the natural logarithm of n. To handle the special case of a weight of 0, it is stipulated that when the weight is 0, the result of multiplying the weight by its natural logarithm is 0. The value of information entropy ranges from 0 to 1. When all samples in the j-th branch channel have the same evaluation value, it indicates that this branch channel provides very limited information, with the greatest uncertainty, and the information entropy reaches its maximum value of 1. Conversely, when the evaluation value of a sample is completely different from other samples, it indicates that this branch channel provides more information, with less uncertainty, and the information entropy is close to 0.

[0076] The difference coefficient reflects the amount of information provided by the j-th branch channel. Subtracting the information entropy of the j-th branch channel from 1 gives the difference coefficient for that branch channel. A larger difference coefficient indicates a smaller information entropy and more information provided by that branch channel, making it more important in assessing the operational efficiency of cold storage facilities.

[0077] Finally, the weight allocation for each branch channel in the evaluation is determined. Specifically, the sum of the difference coefficients for all branch channels is calculated first. Then, the difference coefficient of the j-th branch channel is divided by this sum, and the result is the weight allocation coefficient for the j-th branch channel. Through this calculation, the weight allocation coefficients for each branch channel used to evaluate operational efficiency indicators are obtained, and these coefficients are recorded sequentially to form a set of branch channel weight allocation coefficients. Each coefficient in this set reflects the importance of the corresponding branch channel in evaluating the operational efficiency of the cold storage.

[0078] Step S4312: Based on the branch channel weight allocation coefficients and the operating efficiency index, evaluate the output accuracy of the branch channel set and determine the index branch channel fusion coefficient set. Specifically, the branch channel weight allocation coefficients are objective weights determined based on information entropy, reflecting the amount of information provided by each branch channel. However, in practical applications, the output accuracy of different branch channels varies. Branch channels with higher output accuracy have more reliable evaluation results and should be given greater weight during fusion. Therefore, by combining the branch channel weight allocation coefficients and output accuracy, the index branch channel fusion coefficients can be determined more reasonably.

[0079] First, to evaluate the output accuracy of each operational efficiency indicator for each branch channel, the repeated measures method can be used. Specifically, the same set of samples is selected, and the j-th branch channel is used to evaluate this set of samples multiple times. Each evaluation yields a result, and then the variance or standard deviation of these evaluation results is calculated. Variance or standard deviation is an indicator used to measure the dispersion of data; the smaller the variance or standard deviation, the closer the results of multiple evaluations are, indicating higher output accuracy for this branch channel. The output accuracy value is specified to range from 0 to 1; a larger value indicates higher output accuracy for this branch channel.

[0080] To incorporate output precision into the determination of the fusion coefficients, a precision adjustment factor can be defined. One way to define it is to first find the maximum output precision among all branch channels (a total of m branch channels), that is, the output precision of the branch channel with the highest output precision, and then divide the output precision of the j-th branch channel by this maximum value. After this processing, the value of the precision adjustment factor is normalized to the range of 0 to 1, and is adjusted based on the branch channel with the highest output precision.

[0081] Finally, the fusion coefficient of the index branch channels is calculated. To comprehensively consider the influence of the branch channel weight allocation coefficient and the precision adjustment factor on the fusion coefficient, a weighted average method is used to calculate the index branch channel fusion coefficient. Specifically, an adjustment parameter is introduced, with a value ranging from 0 to 1. This adjustment parameter balances the influence of the branch channel weight allocation coefficient and the output precision on the fusion coefficient. When the value of the adjustment parameter is set close to 1, the calculated fusion coefficient is more influenced by the branch channel weight allocation coefficient; conversely, if the value of the adjustment parameter is set close to 0, the fusion coefficient is more influenced by the precision adjustment factor. In practical applications, a suitable adjustment parameter value can be determined through experiments and analysis. For example, after a series of experiments and analyses, if the value of the adjustment parameter is determined to be 0.7, then the fusion coefficient for each branch channel j can be calculated using this method. The fusion coefficients of all branch channels are recorded sequentially to form the index branch channel fusion coefficient set.

[0082] This implementation method uses the entropy weighting method to determine the weight allocation coefficients of the branch channels, avoiding the interference of human factors in subjective weighting methods, making the weight allocation more objective and scientific. At the same time, by comprehensively considering the output accuracy, the fusion coefficients are further adjusted, which can more accurately reflect the actual role of each branch channel in the evaluation and improve the reliability of the evaluation results.

[0083] This application's embodiments employ techniques such as collecting multi-dimensional data streams of cold storage operation, obtaining historical operation datasets of cold storage, extracting relevant indicators and performing dimensionality reduction analysis to construct a set of energy consumption and operational efficiency indicators, using the aforementioned indicator set to evaluate and classify the historical dataset, obtaining a cold storage energy consumption and operational efficiency indicator dataset, mapping and evaluating the above two indicator datasets, building an evaluation channel, and evaluating the multi-dimensional data streams of cold storage operation to determine the operational efficiency evaluation results. These techniques solve the technical problems of insufficient comprehensiveness and accuracy in existing cold storage operational efficiency evaluations, achieving a comprehensive and accurate technical effect of evaluating cold storage operational efficiency.

[0084] In the above text, refer to Figure 1 This paper describes in detail a method for evaluating the operating efficiency of cold storage facilities by combining energy consumption data analysis according to an embodiment of the present invention. Next, we will refer to... Figure 2 This invention describes a cold storage operation efficiency evaluation system that combines energy consumption data analysis according to an embodiment of the present invention.

[0085] A cold storage operation efficiency evaluation system combining energy consumption data analysis, according to an embodiment of the present invention, addresses the technical problems of insufficient comprehensiveness and accuracy in existing cold storage operation efficiency evaluations, achieving a comprehensive and accurate assessment of cold storage operation efficiency. The system includes: a multi-dimensional data stream acquisition module 10 for cold storage operation, an indicator set construction module 20, an indicator dataset acquisition module 30, and an efficiency evaluation module 40.

[0086] The system includes a multi-dimensional data stream acquisition module 10 for acquiring multi-dimensional data streams of cold storage operation, including power consumption data, refrigerant flow data, temperature and humidity data, and pressure data; an indicator set construction module 20 for acquiring historical cold storage operation datasets, extracting related indicators and performing dimensionality reduction analysis on the historical cold storage operation datasets to construct energy consumption data indicator sets and cold storage operation efficiency indicator sets; an indicator dataset acquisition module 30 for evaluating and classifying the historical cold storage operation datasets using the energy consumption data indicator sets and cold storage operation efficiency indicator sets to obtain cold storage energy consumption indicator datasets and cold storage operation efficiency indicator datasets; and an efficiency evaluation module 40 for mapping and evaluating the cold storage energy consumption indicator datasets and cold storage operation efficiency indicator datasets to build a cold storage operation efficiency evaluation channel, and evaluating the efficiency of the multi-dimensional data streams of cold storage operation based on the cold storage operation efficiency evaluation channel to determine the cold storage operation efficiency evaluation result.

[0087] The detailed description of the specific configuration of the indicator set construction module 20 is explained as follows: As mentioned above, to construct an energy consumption data indicator set and a cold storage operation efficiency indicator set, the indicator set construction module 20 may further include: a data cleaning unit for cleaning the historical operation dataset of the cold storage and arranging it according to time sequence information to obtain a cold storage sequence operation dataset; an association indicator extraction unit for extracting association indicators from the cold storage sequence operation dataset according to the cold storage operation business logic to obtain an energy consumption association indicator set and an operation efficiency association indicator set; a cluster analysis unit for performing cluster analysis on the energy consumption association indicator set and the operation efficiency association indicator set respectively to obtain an energy consumption indicator cluster set and an operation efficiency indicator cluster set; and an indicator dimensionality reduction analysis unit for performing indicator dimensionality reduction analysis based on the energy consumption indicator cluster set and the operation efficiency indicator cluster set to obtain an energy consumption data indicator set and a cold storage operation efficiency indicator set.

[0088] The clustering analysis unit, which obtains the energy consumption index cluster set and the operation efficiency index cluster set, may further include: a hierarchy partitioning subunit for partitioning the energy consumption index cluster set and the operation efficiency index cluster set into hierarchical relationships to obtain energy consumption index hierarchy information and operation efficiency index hierarchy information; a cascade analysis subunit for performing cascade analysis on the energy consumption index cluster set and the operation efficiency index cluster set according to the energy consumption index hierarchy information and the operation efficiency index hierarchy information to obtain an energy consumption index cascade tree and an operation efficiency index cascade tree; an index quantity determination subunit for determining the number of first-level energy consumption indicators and the number of first-level operation efficiency indicators based on the energy consumption index cascade tree and the operation efficiency index cascade tree; and a K-means clustering analysis subunit for performing K-means clustering analysis on the energy consumption index cluster set and the operation efficiency index cluster set using the number of first-level energy consumption indicators and the number of first-level operation efficiency indicators to obtain the energy consumption index cluster set and the operation efficiency index cluster set.

[0089] The energy consumption data index set and cold storage operation efficiency index set acquisition unit may further include: a decision importance analysis subunit for performing decision importance analysis on each index cluster in the energy consumption index cluster set and the operation efficiency index cluster set to obtain an energy consumption index cluster decision factor set and an operation efficiency index cluster decision factor set; an index cluster dimensionality reduction coefficient set determination subunit for determining the energy consumption index cluster dimensionality reduction coefficient set and the operation efficiency index cluster dimensionality reduction coefficient set based on the energy consumption index cluster decision factor set and the operation efficiency index cluster decision factor set; and an index dimensionality reduction processing subunit for performing index dimensionality reduction processing on the energy consumption index cluster set and the operation efficiency index cluster set based on the energy consumption index cluster dimensionality reduction coefficient set and the operation efficiency index cluster dimensionality reduction coefficient set to determine the energy consumption data index set and the cold storage operation efficiency index set.

[0090] The determination of the energy consumption data index set and the cold storage operation efficiency index set, the index dimensionality reduction processing subunit may further include: an index-cluster center distance set acquisition component for obtaining the energy consumption index-cluster center distance set and the operation efficiency index-cluster center distance set based on the energy consumption index cluster set and the operation efficiency index cluster set; a dimensionality reduction distance analysis component for performing dimensionality reduction distance analysis on the energy consumption index-cluster center distance set and the operation efficiency index-cluster center distance set based on the energy consumption index cluster dimensionality reduction coefficient set and the operation efficiency index cluster dimensionality reduction coefficient set, to determine the energy consumption index cluster dimensionality reduction distance set and the operation efficiency index cluster dimensionality reduction distance set; and an index dimensionality reduction processing component for performing index dimensionality reduction processing on the energy consumption index cluster set and the operation efficiency index cluster set using the energy consumption index cluster dimensionality reduction distance set and the operation efficiency index cluster dimensionality reduction distance set, to determine the energy consumption data index set and the cold storage operation efficiency index set.

[0091] The detailed description of the specific configuration of the efficiency evaluation module 40 is as follows: As mentioned above, to build a cold storage operation efficiency evaluation channel, the efficiency evaluation module 40 may further include: an association mapping unit for associating each efficiency index in the cold storage operation efficiency index dataset with the cold storage energy consumption index dataset to obtain an operation efficiency index-associative energy consumption index dataset; an efficiency evaluation training unit for performing efficiency evaluation training on the operation efficiency index-associative energy consumption index dataset to generate an operation efficiency index evaluation branch channel set; and a serial fusion unit for serially fusion of the operation efficiency index evaluation branch channel set to obtain a cold storage operation efficiency evaluation channel.

[0092] The efficiency evaluation training unit, which generates the operational efficiency index evaluation branch channel set, may further include: a branch index channel network structure selection subunit for selecting a branch index channel network structure based on the characteristic information of the operational efficiency index-related energy consumption index dataset; and an efficiency evaluation training subunit for performing efficiency evaluation training on the operational efficiency index-related energy consumption index dataset using the branch index channel network structure to generate the operational efficiency index evaluation branch channel set.

[0093] The cold storage operation efficiency evaluation channel acquisition unit may further include: a fusion coefficient evaluation subunit for evaluating the fusion coefficient of the operation efficiency index evaluation branch channel set to determine the index branch channel fusion coefficient set; and a series fusion subunit for performing series fusion of the operation efficiency index evaluation branch channel set based on the index branch channel fusion coefficient set to obtain the cold storage operation efficiency evaluation channel.

[0094] The determination of the indicator branch channel fusion coefficient set, the fusion coefficient evaluation subunit may further include: an entropy weight calculation component for performing entropy weight calculation on the operational efficiency indicator evaluation branch channel set to determine the branch channel weight allocation coefficient; and an indicator branch channel fusion coefficient set determination component for determining the indicator branch channel fusion coefficient set based on the branch channel weight allocation coefficient and the output accuracy of the operational efficiency indicator evaluation branch channel set.

[0095] The cold storage operation efficiency evaluation system combining energy consumption data analysis provided in this embodiment of the invention can execute the cold storage operation efficiency evaluation method combining energy consumption data analysis provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0096] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0097] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A cold storage operation efficiency evaluation method combined with energy consumption data analysis, characterized in that, The method comprises: Collecting cold storage operation multidimensional data flow, the cold storage operation multidimensional data flow comprising power consumption data, refrigerant flow data, temperature and humidity data, and pressure data; Obtaining a cold storage historical operation data set, performing associated index extraction and dimension reduction analysis on the cold storage historical operation data set, and constructing an energy consumption data index set and a cold storage operation efficiency index set; Using the energy consumption data index set and the cold storage operation efficiency index set to evaluate and classify the cold storage historical operation data set, obtaining a cold storage energy consumption index data set and a cold storage operation efficiency index data set; Mapping and evaluating training the cold storage energy consumption index data set and the cold storage operation efficiency index data set, building a cold storage operation efficiency evaluation channel, and based on the cold storage operation efficiency evaluation channel, performing efficiency evaluation on the cold storage operation multidimensional data flow to determine a cold storage operation efficiency evaluation result.

2. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 1, characterized in that, The construction of the energy consumption data index set and the cold storage operation efficiency index set comprises: Performing data cleaning processing on the cold storage historical operation data set and arranging it according to time sequence information to obtain a cold storage sequence operation data set; Performing associated index extraction on the cold storage sequence operation data set according to cold storage operation business logic to obtain an energy consumption associated index set and an operation efficiency associated index set; Performing clustering analysis on the energy consumption associated index set and the operation efficiency associated index set respectively to obtain an energy consumption index cluster set and an operation efficiency index cluster set; Performing index dimension reduction analysis based on the energy consumption index cluster set and the operation efficiency index cluster set to obtain an energy consumption data index set and a cold storage operation efficiency index set.

3. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 2, characterized in that, The obtaining of the energy consumption index cluster set and the operation efficiency index cluster set comprises: Performing associated level division on the energy consumption associated index set and the operation efficiency associated index set respectively to obtain energy consumption index level information and operation efficiency index level information; Performing cascade analysis on the energy consumption associated index set and the operation efficiency associated index set according to the energy consumption index level information and the operation efficiency index level information to obtain an energy consumption index cascade tree and an operation efficiency index cascade tree; According to the energy consumption index cascade tree and the operation efficiency index cascade tree, determining a first-level energy consumption index number and a first-level operation efficiency index number; Using the first-level energy consumption index number and the first-level operation efficiency index number to perform K-means clustering analysis on the energy consumption associated index set and the operation efficiency associated index set respectively to obtain the energy consumption index cluster set and the operation efficiency index cluster set.

4. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 2, characterized in that, The obtaining of the energy consumption data index set and the cold storage operation efficiency index set comprises: Performing decision importance analysis on each index cluster in the energy consumption index cluster set and the operation efficiency index cluster set to obtain an energy consumption index cluster decision factor set and an operation efficiency index cluster decision factor set; According to the energy consumption index cluster decision factor set and the operation efficiency index cluster decision factor set, determining an energy consumption index cluster dimension reduction coefficient set and an operation efficiency index cluster dimension reduction coefficient set; Based on the energy consumption index cluster dimension reduction coefficient set and the operation efficiency index cluster dimension reduction coefficient set, performing index dimension reduction processing on the energy consumption index cluster set and the operation efficiency index cluster set to determine an energy consumption data index set and a cold storage operation efficiency index set.

5. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 4, characterized in that, The determined energy consumption data indicator set and cold storage operation efficiency indicator set comprise: According to the energy consumption indicator cluster set and the operation efficiency indicator cluster set, an energy consumption indicator-cluster center distance set and an operation efficiency indicator-cluster center distance set are obtained; Based on the energy consumption indicator cluster dimension reduction coefficient set and the operation efficiency indicator cluster dimension reduction coefficient set, dimension reduction distance analysis is performed on the energy consumption indicator-cluster center distance set and the operation efficiency indicator-cluster center distance set to determine an energy consumption indicator cluster dimension reduction distance set and an operation efficiency indicator cluster dimension reduction distance set; The energy consumption indicator cluster set and the operation efficiency indicator cluster set are subjected to indicator dimension reduction processing by using the energy consumption indicator cluster dimension reduction distance set and the operation efficiency indicator cluster dimension reduction distance set to determine the energy consumption data indicator set and the cold storage operation efficiency indicator set.

6. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 1, characterized in that, The cold storage operation efficiency evaluation channel is built by: Each efficiency indicator in the cold storage operation efficiency indicator data set is associated and mapped with the cold storage energy consumption indicator data set to obtain an operation efficiency indicator-associated energy consumption indicator data set; The operation efficiency indicator-associated energy consumption indicator data set is subjected to efficiency evaluation training respectively to generate an operation efficiency indicator evaluation branch channel set; The operation efficiency indicator evaluation branch channel set is serially fused to obtain the cold storage operation efficiency evaluation channel.

7. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 6, characterized in that, The operation efficiency indicator evaluation branch channel set is generated by: According to the characteristic information of the operation efficiency indicator-associated energy consumption indicator data set, a branch indicator channel network structure is selected; The operation efficiency indicator-associated energy consumption indicator data set is subjected to efficiency evaluation training respectively by using the branch indicator channel network structure to generate the operation efficiency indicator evaluation branch channel set.

8. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 6, characterized in that, The cold storage operation efficiency evaluation channel is obtained by: The operation efficiency indicator evaluation branch channel set is subjected to fusion coefficient evaluation to determine an indicator branch channel fusion coefficient set; Based on the indicator branch channel fusion coefficient set, the operation efficiency indicator evaluation branch channel set is serially fused to obtain the cold storage operation efficiency evaluation channel.

9. The method for evaluating the operation efficiency of a cold storage in combination with energy consumption data analysis according to claim 8, characterized in that, The indicator branch channel fusion coefficient set is determined by: The operation efficiency indicator evaluation branch channel set is subjected to entropy weight calculation to determine a branch channel weight distribution coefficient; Based on the branch channel weight distribution coefficient and the output accuracy of the operation efficiency indicator evaluation branch channel set, the indicator branch channel fusion coefficient set is determined.

10. A cold storage operation efficiency evaluation system combined with energy consumption data analysis, characterized in that, The system is used to implement the cold storage operation efficiency evaluation method combined with energy consumption data analysis according to any one of claims 1-9, and the system comprises: A cold storage operation multi-dimensional data stream acquisition module is configured to acquire cold storage operation multi-dimensional data stream, wherein the cold storage operation multi-dimensional data stream comprises consumed power data, refrigerant flow data, temperature and humidity data, and pressure data; An indicator set construction module is configured to obtain a cold storage historical operation data set, perform associated indicator extraction and dimension reduction analysis on the cold storage historical operation data set, and construct an energy consumption data indicator set and a cold storage operation efficiency indicator set; and An operation efficiency evaluation channel is built by: Each efficiency indicator in the cold storage operation efficiency indicator data set is associated and mapped with the cold storage energy consumption indicator data set to obtain an operation efficiency indicator-associated energy consumption indicator data set; The operation efficiency indicator-associated energy consumption indicator data set is subjected to efficiency evaluation training respectively to generate an operation efficiency indicator evaluation branch channel set; The operation efficiency indicator evaluation branch channel set is serially fused to obtain the cold storage operation efficiency evaluation channel. The operation efficiency indicator evaluation branch channel set is generated by: According to the characteristic information of the operation efficiency indicator-associated energy consumption indicator data set, a branch indicator channel network structure is selected; The operation efficiency indicator-associated energy consumption indicator data set is subjected to efficiency evaluation training respectively by using the branch indicator channel network structure to generate the operation efficiency indicator evaluation branch channel set. The cold storage operation efficiency evaluation channel is obtained by: The operation efficiency indicator evaluation branch channel set is subjected to fusion coefficient evaluation to determine an indicator branch channel fusion coefficient set; Based on the indicator branch channel fusion coefficient set, the operation efficiency indicator evaluation branch channel set is serially fused to obtain the cold storage operation efficiency evaluation channel. The indicator branch channel fusion coefficient set is determined by: The operation efficiency indicator evaluation branch channel set is subjected to entropy weight calculation to determine a branch channel weight distribution coefficient; Based on the branch channel weight distribution coefficient and the output accuracy of the operation efficiency indicator evaluation branch channel set, the indicator branch channel fusion coefficient set is determined. An index data set acquisition module is configured to evaluate and classify the cold storage historical operation data set by using the energy consumption data index set and the cold storage operation efficiency index set, so as to obtain a cold storage energy consumption index data set and a cold storage operation efficiency index data set; An efficiency evaluation module is configured to map and evaluate and train the cold storage energy consumption index data set and the cold storage operation efficiency index data set, build a cold storage operation efficiency evaluation channel, and evaluate the efficiency of the cold storage operation multi-dimensional data flow based on the cold storage operation efficiency evaluation channel to determine a cold storage operation efficiency evaluation result.

Images