An intelligent refrigeration host energy efficiency monitoring system

Abstract

The present invention relates to the field of intelligent monitoring technology, specifically to an intelligent refrigeration host energy efficiency monitoring system, the system includes a data acquisition and processing module, a time series analysis module, an energy efficiency control module and an abnormal response module, based on temperature sensor data, collects time series data of the refrigeration host under different operating states, and corrects the time deviation value and data volume normalization. The present invention accurately processes time series data through a dynamic time bending algorithm, and can monitor the operating efficiency and state changes of the refrigeration host in more detail, especially showing great advantages in identifying subtle efficiency changes and abnormal patterns. Combined with the intelligent classification of decision trees, the system can respond quickly and accurately when potential problems are discovered, thereby giving early warnings or adjusting strategies. Through early warning mechanisms and immediate parameter adjustments, the reliability and efficiency of the refrigeration host are significantly improved, and energy consumption and operating costs are effectively reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to an intelligent refrigeration host energy efficiency monitoring system. Background Art

[0002] Intelligent monitoring technology mainly uses sensors, data analysis software, and artificial intelligence decision support to monitor and control various environmental and mechanical systems. Intelligent monitoring technology can provide real-time data feedback, optimize operational efficiency, reduce energy waste, and improve the overall performance of the system.

[0003] Among them, the intelligent refrigeration host energy efficiency monitoring system optimizes energy use and improves system efficiency by real-time monitoring and adjusting the performance of the refrigeration host. It uses sensors to monitor key parameters such as temperature, pressure and current, and analyzes data through AI algorithms to adjust operating status and reduce energy consumption. Its main use is to achieve energy efficiency management in commercial and industrial environments, ensuring that refrigeration equipment minimizes power consumption and maintenance costs while providing necessary cooling, thereby supporting environmental sustainability and reducing operating expenses.

[0004] Existing technologies lack dynamic adaptability in data analysis and are unable to promptly identify signs of reduced efficiency or failure when faced with complex or nonlinear data patterns. This analytical limitation can cause the refrigeration host to continue operating without timely adjustment of operating parameters, increasing the risk of energy efficiency loss and equipment damage. The lack of data processing tools and rapid response mechanisms also causes problems for existing technologies in maintaining system stability and maximizing energy efficiency, especially in an operating environment where demand is constantly changing, increasing operating costs and the risk of failure. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent refrigeration host energy efficiency monitoring system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: an intelligent refrigeration host energy efficiency monitoring system, the system comprising:

[0007] The data acquisition and processing module collects the time series data of the refrigeration host under different operating conditions based on the temperature sensor data, corrects the time deviation value and normalizes the data volume, extracts key operating characteristics, and constructs feature integrated data;

[0008] The time series analysis module uses a dynamic time warping algorithm to perform similarity calculation on the time series based on the feature integrated data, identifies the efficiency changes and abnormal patterns during the operation of the refrigeration host, and obtains an abnormality identification index;

[0009] The energy efficiency control module adjusts the key operating parameters of the refrigeration host, including compressor speed and temperature settings, based on the abnormal identification index, optimizes the performance and energy efficiency of the refrigeration host, and obtains parameter control information;

[0010] The abnormal response module uses the decision tree algorithm to intelligently classify the current operating mode based on the parameter control information and the historical performance data of the refrigeration host to determine whether to trigger an early warning. If an operating abnormality is detected, automatic adjustment is performed to generate a response decision record.

[0011] The present invention is improved in that the steps of collecting the time series data are specifically as follows:

[0012] Based on the temperature sensor data, monitor the temperature data of the refrigeration host during startup, operation and shutdown, record the temperature values ​​with timestamps, and generate the original time series data set;

[0013] Based on the original time series data set, the formula is adopted:

[0014] ,

[0015] Calculate and correct the time and temperature errors in the data to obtain a processed time series data set ,in, represents the data points in the original dataset, is the timestamp error, , and They are the average, maximum and minimum temperatures recorded in history.

[0016] The present invention is improved in that the step of acquiring the feature integration data is specifically as follows:

[0017] Extracting key operating characteristics of the refrigeration host, including the highest temperature, the lowest temperature, the average temperature and the temperature change frequency in each state cycle, to obtain a preliminary characteristic data set;

[0018] Based on the preliminary feature data set, the formula is adopted:

[0019] ,

[0020] Balance weights and principal component contributions to build feature integration data ,in, represents the i-th principal component, is the weight of the ith principal component, is the absolute value of the ith weight, and n is the total number of principal components.

[0021] The present invention is improved in that the steps of similarity calculation are specifically as follows:

[0022] Based on the feature integrated data, a time series including key performance indicators is selected to obtain a data set to be analyzed;

[0023] Based on the data set to be analyzed, the dynamic time warping algorithm is used to calculate the similarity between time series using the formula:

[0024] ,

[0025] Generate a similarity matrix, where represents the cumulative minimum distance between the i-th element and the j-th element, is the value of the i-th element in the time series s, is the value of the jth element in time series t, is a time series and The square of the difference between and is the penalty term in the adjustment sequence matching step.

[0026] The present invention is improved in that the step of obtaining the abnormal identification index is specifically as follows:

[0027] Identify the time points of deviation from normal performance, analyze potential efficiency changes and anomalies, and obtain a preliminary anomaly candidate list;

[0028] Based on the preliminary abnormality candidate table, the criticality of each candidate abnormality is evaluated using the formula:

[0029] ,

[0030] Determine the efficiency changes and abnormal patterns during the operation of the refrigeration host and obtain abnormal identification indicators ,in, is the similarity score of the ith candidate anomaly pattern, is the average of the similarity scores, is the weight assigned to the ith data point, is the number of candidate anomaly patterns.

[0031] The present invention is improved in that the step of obtaining the parameter control information is specifically as follows:

[0032] Based on the abnormal identification index, real-time operation data of the refrigeration host is collected, and data changes within the period are summarized to obtain a real-time operation data summary;

[0033] Based on the real-time operation data summary, the formula is adopted:

[0034] ,

[0035] Calculate the coefficient of variation to obtain the statistical volatility of the operating parameters, where is the variance of the data, is a single data point, is the mean of the data points, is the total number of data points;

[0036] According to the statistical volatility of the operating parameters, the adjusted target compressor speed and temperature settings are determined to optimize energy efficiency and reduce energy consumption, thereby obtaining parameter control information.

[0037] The present invention is improved in that the classification step of the current operation mode is specifically as follows:

[0038] Based on the parameter control information, including the adjusted speed and temperature of the compressor, record the data affecting the performance of the refrigeration host and generate a data control log;

[0039] Based on the data control log, combined with the historical performance data of the refrigeration host, the difference and correlation are calculated using the formula:

[0040] ,

[0041] Get a complete performance data set, where is the performance score, and Represent the standard deviation of current and historical data, and is the adjustment factor;

[0042] The complete performance data set is analyzed using a decision tree algorithm, the operation modes are intelligently classified according to the performance scores and data representations, the operation status is determined, and an intelligent classification result is generated.

[0043] The present invention is improved in that the step of obtaining the response decision record is specifically as follows:

[0044] Organize current and historical performance data, mark each piece of data with the corresponding timestamp and impact indicator, and obtain an integrated performance set;

[0045] Based on the integrated performance set, the formula is adopted:

[0046] ,

[0047] Calculate the performance change between the current data point and the historical data point, and automatically trigger the adjustment process or alarm according to the change value to obtain the response decision record, where: is the percentage change in performance, Represents the current performance data point, Indicates the corresponding historical data point.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are:

[0049] In the present invention, the dynamic time bending algorithm is used to accurately process time series data, so that the operating efficiency and state changes of the refrigeration host can be monitored more carefully, especially showing great advantages in identifying subtle efficiency changes and abnormal patterns. Combined with the intelligent classification of the decision tree, the system can respond quickly and accurately when potential problems are found, thereby providing early warning or adjusting strategies. Through the early warning mechanism and real-time parameter adjustment, the reliability and efficiency of the refrigeration host are significantly improved, ensuring optimal performance under various operating conditions, and effectively reducing energy consumption and operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a system flow chart of the present invention;

[0051] Figure 2 It is a flow chart for collecting time series data in the present invention;

[0052] Figure 3 This is a flow chart for obtaining feature integration data in the present invention;

[0053] Figure 4 This is a flow chart for similarity calculation in the present invention;

[0054] Figure 5 This is a flow chart for obtaining abnormal identification indicators in the present invention;

[0055] Figure 6 This is a flow chart for obtaining parameter control information in the present invention;

[0056] Figure 7 is a classification flow chart of the current operation mode in the present invention;

[0057] Figure 8 The present invention is a flowchart for obtaining a response decision record. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0060] Example:

[0061] See also Figure 1 The present invention provides a technical solution: an intelligent refrigeration host energy efficiency monitoring system comprises:

[0062] The data acquisition and processing module collects the time series data of the refrigeration host under different operating conditions based on the temperature sensor data, corrects the time deviation value and normalizes the data volume, extracts key operating characteristics, and constructs feature integrated data;

[0063] The time series analysis module uses the dynamic time warping algorithm to calculate the similarity of the time series based on the feature integration data, identify the efficiency changes and abnormal patterns during the operation of the refrigeration host, and obtain the abnormal identification index;

[0064] The energy efficiency control module adjusts the key operating parameters of the refrigeration host, including compressor speed and temperature settings, based on abnormal identification indicators, optimizes the performance and energy efficiency of the refrigeration host, and obtains parameter control information;

[0065] The abnormal response module uses the decision tree algorithm to intelligently classify the current operating mode based on parameter control information and the historical performance data of the refrigeration host to determine whether to trigger an early warning. If an operating abnormality is detected, automatic adjustment is performed to generate a response decision record.

[0066] For abnormal automatic adjustments, there are the following, etc. If the sensor detects that the compressor temperature rises abnormally, it may be caused by overload or insufficient refrigerant. The system will automatically reduce the compressor speed, or prompt to add or check the refrigerant to ensure that the compressor is not damaged due to overheating;

[0067] When the energy efficiency analysis shows that the refrigeration efficiency is reduced, the system will automatically adjust the pressure and temperature parameters in the refrigeration cycle to optimize the heat exchange process. If the problem persists, it can prompt a mechanical inspection, including cleaning or repair of the condenser and evaporator.

[0068] When the external ambient temperature changes dramatically (such as heat waves or cold snaps), if the refrigeration host fails to effectively adjust to cope with such changes, the system will automatically recalculate the required cooling load and adjust the operating mode, including changing the operating frequency of the compressor and adjusting the air flow control door or fan speed to maintain a stable indoor environment;

[0069] If an abnormal drop in refrigerant pressure is detected, the system will immediately start the leak detection program, reduce or suspend the operation of the compressor, and issue an alarm through the system interface. At the same time, the system automatically starts the leak plugging program to reduce the loss of refrigerant and prompt emergency maintenance.

[0070] Feature integration data includes operating time points, temperature readings, and adjusted data indicators. Abnormal identification indicators include operating deviation rate, mode abnormality index, and critical time window. Parameter control information includes compressor speed adjustment parameters and temperature setting adjustment parameters. Response decision records include alarm levels, adjustment plans, and effect comparison results.

[0071] See also Figure 2 , the specific steps for collecting time series data are:

[0072] Based on the temperature sensor data, monitor the temperature data of the refrigeration host during startup, operation and shutdown, record the temperature values ​​with timestamps, and generate the original time series data set;

[0073] When configuring temperature sensors at key locations of the refrigeration host, the mutual influence of the installation location of the temperature sensor and environmental factors is considered to ensure that the data obtained can truly reflect the operating status of the machine. The temperature data is monitored in a continuous manner, with data recorded at least once a minute in each state. During the collection period, the stability of the equipment and the reliability of data transmission are tested. The recorded data include timestamps and corresponding temperature values. The parameters involved in the specific process of data collection include time intervals, temperature value ranges, sensor accuracy, etc. The parameters are set according to the actual operating conditions and equipment specifications of the refrigeration host to ensure that the collected data is both detailed and representative, and the original time series data set is generated.

[0074] Based on the original time series data set, the formula is used:

[0075] ,

[0076] Calculate and correct the time and temperature errors in the data to obtain a processed time series data set ,in, represents the data points in the original dataset, is the timestamp error, , and The average, maximum and minimum temperatures in history. represents the square correction effect of the timestamp error, Used to normalize temperature errors and improve the accuracy of data correction;

[0077] One raw data point collected during the monitoring period is , at time , the record should be in So = 1 minute, that is =1, if , , Then the normalized temperature error term is:

[0078] ,

[0079] Substituting into the formula, we get: ;

[0080] The results showed that 29.5℃ was obtained through the adjusted cleaning process. Slightly higher than the original reading, it reflects the dual correction of time error and temperature range, making the data closer to the temperature performance of the actual operating status.

[0081] See also Figure 3 , the specific steps for obtaining feature integration data are:

[0082] Extract the key operating characteristics of the refrigeration host, including the maximum temperature, minimum temperature, average temperature and temperature change frequency in each state cycle, and obtain a preliminary feature data set;

[0083] The process of extracting the operating characteristics of the refrigeration host from the time series data set after time deviation correction involves multi-angle analysis of temperature data, covering the determination of the highest temperature, lowest temperature, average temperature and temperature change frequency in each state cycle. The parameters are automatically calculated based on the data points to ensure accurate data extraction. When calculating the average temperature, the sum of the temperatures of all data points is determined and then divided by the number of data points. The highest and lowest temperatures are directly determined by comparing all measured temperature values. Frequency analysis is performed by identifying the regularity and frequency of temperature changes to ensure consistency and accuracy of operations. The generated feature data set provides input for subsequent data analysis and machine learning models.

[0084] Based on the preliminary feature data set, the formula is used:

[0085] ,

[0086] Balance weights and principal component contributions to build feature integration data ,in, represents the i-th principal component, is the weight of the ith principal component, is the absolute value of the i-th weight, used to calculate the sum of weights, and n is the total number of principal components;

[0087] From the principal component analysis, three main components were obtained, =0.8, =0.15, =0.05, and the corresponding weights are set to =0.5, =0.3, =0.2;

[0088] Substitute into the formula to calculate: ;

[0089] The results show that after comprehensively considering the contribution of each principal component, the obtained feature integrated data value is 0.455, reflecting that the key features of the refrigerator operation status are effectively extracted and weighted, providing a high-quality data basis for subsequent performance evaluation and fault prediction.

[0090] See also Figure 4 , the specific steps for similarity calculation are:

[0091] Based on the feature integrated data, the time series including key performance indicators, including the temperature and pressure of the refrigeration host, are selected to obtain the data set to be analyzed;

[0092] Key performance indicators, such as temperature and pressure data, are selected from the feature integration data. The time series data involved are used for subsequent dynamic time warping algorithm processing. Based on this data set, preprocessing operations are performed, including data standardization and missing value processing, to ensure data quality and prepare sufficient time series samples for similarity calculation. Preprocessing includes data interpolation and removal of outliers to ensure the consistency and reliability of the data set, so that the subsequent DTW algorithm application can more accurately reflect the dynamic similarity between time series.

[0093] Based on the data set to be analyzed, the dynamic time warping algorithm is used to calculate the similarity between time series using the formula:

[0094] ,

[0095] Generate a similarity matrix, where is an element of the dynamic time-warping distance matrix between time series s and t, representing the cumulative minimum distance between the ith element and the jth element, is the value of the i-th element in the time series s, is the value of the jth element in time series t, is a time series and The square of the difference between and is the penalty term in the adjustment sequence matching step, which is used to align time series of different lengths. , and is the dynamic time warping distance calculated in the previous step of dynamic programming, representing the matching points in the upper left corner, above, and to the left, respectively, for the current point Calculation of

[0096] There are two time series s and t, and the values ​​at time points i and j are =2 and =3, set adjustment items =1 and =1.

[0097] Calculate the squared difference of the distance between points:

[0098] ,

[0099] Recursively calculate the three estimated path costs and select the minimum value as the current :

[0100] From the upper left (matching point and , (set up );

[0101] from above (skip point i or increase i), (set up );

[0102] From the left (skip point j or increase j), (set up );

[0103] Select the minimum cost path to update the current :

[0104] ,

[0105] get , indicating that at time points i and j, the similarity score of sequences s and t is 1. This low score indicates a high similarity between the two points, which helps to identify and analyze important patterns and abnormal behaviors in time series in subsequent steps.

[0106] See also Figure 5 ,The specific steps for obtaining abnormal identification indicators are:

[0107] Identify the time points of deviation from normal performance, analyze potential efficiency changes and anomalies, and obtain a preliminary anomaly candidate list;

[0108] Based on the preliminary analysis of time series data and abnormal patterns, the key time points that affect the efficiency of the refrigeration host are determined. The data points show behaviors that are significantly different from the normal operating mode. Data processing includes extracting key features from the time series, using statistical methods for analysis, determining the severity of the abnormal patterns, and generating a preliminary abnormality identification candidate list. The candidate list will subsequently undergo an evaluation process to ensure that it can accurately identify and respond to operating problems, providing a basis for the maintenance of the refrigeration system and ensuring efficient and stable system operation.

[0109] Based on the preliminary anomaly candidate table, evaluate the criticality of each candidate anomaly using the formula:

[0110] ,

[0111] Determine the efficiency changes and abnormal patterns during the operation of the refrigeration host and obtain abnormal identification indicators ,in, is the similarity score of the ith candidate anomaly pattern, is the average of the similarity scores, is the weight assigned to the ith data point, is the number of candidate anomaly patterns;

[0112] If the specific value is , , , , the weights are , , , the calculation process is:

[0113] ,

[0114] The calculation results ensure that the degree of deviation of each data point from the average performance can be quantified, thereby identifying potential anomalies. A high value indicates that there is a suspicion of one or more significant performance deviation points.

[0115] See also Figure 6 , the specific steps for obtaining parameter control information are:

[0116] Based on the abnormal identification indicators, the real-time operation data of the refrigeration host is collected, including the compressor speed and temperature, etc., and the data changes within the cycle are summarized to obtain the real-time operation data summary;

[0117] Collect real-time operating data of the refrigeration host, including the speed and temperature information of the compressor, and accumulate data by continuously monitoring parameters during the operating cycle. By accumulating data, the consistency and abnormal fluctuations of the machine operation can be analyzed, and a real-time operating data summary can be generated. The summarized data provides a basis for subsequent analysis, ensuring the integrity of the data and the accuracy of the analysis, and ensuring that the collected data can truly reflect the operating status of the refrigerator.

[0118] Based on the real-time operation data summary, the formula is used:

[0119] ,

[0120] Calculate the coefficient of variation to obtain the statistical volatility of the operating parameters, where is the variance of the data, is a single data point, is the mean of the data points, is the total number of data points;

[0121] If we have the following data points, =80, =85, =78, average =81, total number of data =3, calculate the square of the deviation of each point:

[0122] ,

[0123] The sum is: ,

[0124] The variance is calculated as: ;

[0125] The results show that the compressor operating parameters fluctuated moderately during the investigation period, and the variance value of 8.67 reflects the degree of dispersion of the data, which helps to determine whether the compressor operating parameters need to be further adjusted to optimize performance and energy efficiency.

[0126] According to the statistical volatility of operating parameters, the target compressor speed and temperature settings are determined to optimize energy efficiency and reduce energy consumption, and parameter control information is obtained;

[0127] After completing the analysis of the volatility of the operating parameters, the target compressor speed and temperature settings are optimized in combination with the performance standards and safety requirements of the equipment to ensure that the refrigeration host maintains efficient operation without causing equipment overload or damage due to improper parameter settings. By adjusting the operating parameters of the refrigeration host, the energy efficiency and performance of the machine are effectively improved, while ensuring the safety of equipment operation, thereby obtaining parameter adjustment information and providing key data for subsequent operation monitoring and maintenance;

[0128] First, the system regularly collects the operating speed and temperature data of the compressor. Assuming that the data is measured three times a day in a week, the data obtained is:

[0129] Speed ​​(RPM): [1450, 1500, 1480, 1470, 1490, 1460, 1500];

[0130] Temperature (℃): [72, 70, 73, 74, 71, 72, 70];

[0131] Calculate the mean and variance of the data, taking speed as an example:

[0132] Average speed =(1450+1500+1480+1470+1490+1460+1500) / 7≈1407RPM;

[0133] Calculate the variance. Assume that the variance is 326 and adjust the running parameters according to the variance:

[0134] Use variance to determine the volatility of the data. Higher variance indicates larger fluctuations. You need to adjust the settings to reduce this fluctuation to ensure more stable equipment operation. If the variance exceeds the threshold (set the threshold to 300), consider reducing the maximum and minimum speed settings and narrowing the adjustment range. For example, the upper and lower limits of the speed can be adjusted to 1400RPM to 1420RPM to make the equipment run in a more stable state. Make similar adjustments to the temperature parameters. If the variance of the temperature is also high, you need to optimize the cooling system or adjust the temperature setting range. After the adjustment, continue to monitor the new operating data to evaluate whether the adjustment effect has reduced volatility and improved the overall performance of the equipment, reducing energy consumption and maintenance costs.

[0135] See also Figure 7 , the classification steps of the current operation mode are as follows:

[0136] Based on the parameter control information, including the speed and temperature of the compressor after adjustment, the data affecting the performance of the refrigeration host is recorded and a data control log is generated;

[0137] Based on the current operating parameters and the speed and temperature information of the compressor after adjustment, the data that affects the performance of the refrigeration host is recorded to ensure that each piece of data is accurately marked with the corresponding timestamp and impact indicator. This is completed through real-time monitoring of parameter changes, including the increase or decrease of the compressor speed and the increase or decrease of the temperature. Each adjusted data is recorded in the data log, providing a complete data source for subsequent analysis. The records are stored in the database in time series for subsequent retrieval and analysis. The generated data control log is intended to provide basic data support for the performance evaluation and fault prediction of the refrigeration host. In this way, the maintenance team can quickly locate the estimated problem points and make corresponding adjustments;

[0138] Based on the data control log, combined with the historical performance data of the refrigeration host, the difference and correlation are calculated using the formula:

[0139] ,

[0140] Get a complete performance data set, where is the performance score, and Represent the standard deviation of current and historical data, and is the adjustment factor used to adjust the influence of the standard deviation of the current and historical data sets on the total score;

[0141] If the standard deviation of the current data =0.05, standard deviation of historical data =0.03, adjustment factor =0.6 and =0.4, then the formula is:

[0142] ,

[0143] The results show that the current operating mode has smaller fluctuations than the historical mode, but we still need to pay attention to the potential impact caused by the changes. Further analysis can provide more accurate data support for decision tree classification;

[0144] Use the decision tree algorithm to analyze the complete performance data set, intelligently classify the operation mode according to the performance score and data performance, judge the operation status, and generate intelligent classification results;

[0145] The decision tree algorithm is used to analyze the complete performance data set, which includes the performance score and performance of each data. By identifying the characteristics of various operating modes and intelligently classifying them, the current operating status is judged not only based on the performance score, but also considering the specific performance of the data such as temperature fluctuations, speed changes, etc., and dynamically adjusted according to the actual data performance to determine whether the refrigeration host is in the best operating state or whether maintenance adjustments are required. The intelligent classification results provide an important basis for operational decisions, helping the maintenance team to make accurate response strategies, thereby ensuring the continuous and stable operation of the refrigeration host.

[0146] See also Figure 8 , the specific steps for obtaining the response decision record are:

[0147] Organize current and historical performance data, mark each piece of data with the corresponding timestamp and impact indicator, and obtain an integrated performance set;

[0148] Organize current and historical performance data, mark each piece of data with the corresponding timestamp and influencing indicators, and in this way, create a performance data set. The data set provides a basis for subsequent analysis and decision-making. By marking the timestamp, the performance trend can be tracked, and the clarification of the influencing indicators helps identify which factors are the most critical, ensuring the integrity and reliability of the data. The integrated performance set obtained will be directly used for the next step of intelligent classification analysis.

[0149] Based on the integrated performance set, the formula is adopted:

[0150] ,

[0151] Calculate the performance change between the current data point and the historical data point, and automatically trigger the adjustment process or alarm according to the change value to obtain the response decision record, where: is the percentage change in performance, Represents the current performance data point, Indicates the corresponding historical data point;

[0152] There are the following data, current performance data points is 200, historical data points is 180, the percentage change is calculated as follows:

[0153] ;

[0154] This result shows that the performance is improved by about 11.11%, which helps decision makers understand the actual effect of the adjustment and make further decisions accordingly.

[0155] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent refrigeration host energy efficiency monitoring system, characterized in that: The system comprises: The data acquisition and processing module collects the time series data of the refrigeration host under different operating conditions based on the temperature sensor data, corrects the time deviation value and normalizes the data volume, extracts key operating characteristics, and constructs feature integrated data; The time series analysis module uses a dynamic time warping algorithm to perform similarity calculation on the time series based on the feature integrated data, identifies the efficiency changes and abnormal patterns during the operation of the refrigeration host, and obtains an abnormal identification index; The energy efficiency control module adjusts the key operating parameters of the refrigeration host, including compressor speed and temperature settings, based on the abnormal identification index, optimizes the performance and energy efficiency of the refrigeration host, and obtains parameter control information; The abnormal response module uses the decision tree algorithm to intelligently classify the current operating mode based on the parameter control information and the historical performance data of the refrigeration host to determine whether to trigger an early warning. If an operating abnormality is detected, automatic adjustment is performed to generate a response decision record.

2. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The steps for collecting the time series data are specifically as follows: Based on the temperature sensor data, monitor the temperature data of the refrigeration host during startup, operation and shutdown, record the temperature values ​​with timestamps, and generate the original time series data set; Based on the original time series data set, the formula is adopted: ， Calculate and correct the time and temperature errors in the data to obtain a processed time series data set ,in, represents the data points in the original dataset, is the timestamp error, , and They are the average, maximum and minimum temperatures recorded in history.

3. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The steps for acquiring the feature integration data are specifically as follows: Extracting key operating characteristics of the refrigeration host, including the highest temperature, the lowest temperature, the average temperature and the temperature change frequency in each state cycle, to obtain a preliminary characteristic data set; Based on the preliminary feature data set, the formula is adopted: ， Balance weights and principal component contributions to build feature integration data ,in, represents the i-th principal component, is the weight of the ith principal component, is the absolute value of the ith weight, and n is the total number of principal components.

4. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The steps of similarity calculation are specifically as follows: Based on the feature integrated data, a time series including key performance indicators is selected to obtain a data set to be analyzed; Based on the data set to be analyzed, the dynamic time warping algorithm is used to calculate the similarity between time series using the formula: ， Generate a similarity matrix, where represents the cumulative minimum distance between the i-th element and the j-th element, is the value of the i-th element in the time series s, is the value of the jth element in time series t, is a time series and The square of the difference between and is the penalty term in the adjustment sequence matching step.

5. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The steps for obtaining the abnormal identification index are specifically as follows: Identify the time points of deviation from normal performance, analyze potential efficiency changes and anomalies, and obtain a preliminary anomaly candidate list; Based on the preliminary abnormality candidate table, the criticality of each candidate abnormality is evaluated using the formula: ， Determine the efficiency changes and abnormal patterns during the operation of the refrigeration host and obtain abnormal identification indicators ,in, is the similarity score of the ith candidate anomaly pattern, is the average of the similarity scores, is the weight assigned to the ith data point, is the number of candidate anomaly patterns.

6. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The steps for obtaining the parameter control information are specifically as follows: Based on the abnormal identification index, real-time operation data of the refrigeration host is collected, and data changes within the period are summarized to obtain a real-time operation data summary; Based on the real-time operation data summary, the formula is adopted: ， Calculate the coefficient of variation to obtain the statistical volatility of the operating parameters, where is the variance of the data, is a single data point, is the mean of the data points, is the total number of data points; According to the statistical volatility of the operating parameters, the adjusted target compressor speed and temperature settings are determined to optimize energy efficiency and reduce energy consumption, thereby obtaining parameter control information.

7. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The classification steps of the current operation mode are specifically as follows: Based on the parameter control information, including the adjusted speed and temperature of the compressor, data affecting the performance of the refrigeration host is recorded to generate a data control log; Based on the data control log, combined with the historical performance data of the refrigeration host, the difference and correlation are calculated using the formula: ， Get a complete performance data set, where is the performance score, and Represent the standard deviation of current and historical data, and is the adjustment factor; The complete performance data set is analyzed using a decision tree algorithm, the operation modes are intelligently classified according to the performance scores and data representations, the operation status is determined, and an intelligent classification result is generated.

8. The intelligent refrigeration host energy efficiency monitoring system according to claim 1 is characterized in that: The steps for obtaining the response decision record are specifically as follows: Organize current and historical performance data, mark each piece of data with the corresponding timestamp and impact indicator, and obtain an integrated performance set; Based on the integrated performance set, the formula is adopted: ， Calculate the performance change between the current data point and the historical data point, and automatically trigger the adjustment process or alarm according to the change value to obtain the response decision record, where: is the percentage change in performance, Represents the current performance data point, Indicates the corresponding historical data point.

Images