Analysis Method and System for Cold Storage Efficiency of Cold Storage Tank Based on Multi-Source Data Fusion

Abstract

The present invention relates to the technical field of data fusion, and specifically relates to a method and system for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion. The method includes: obtaining the evaluation index value of a monitoring sensor according to the difference between the temperature sequence to be analyzed of the monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, the correlation between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensor, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor; obtaining the inlet fusion temperature sequence and the outlet fusion temperature sequence according to the evaluation index value, and obtaining the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle according to the inlet fusion temperature sequence and the outlet fusion temperature sequence. Moreover, the present invention can improve the effectiveness and accuracy of the evaluation of the cold storage efficiency of the cold storage tank.

Description

Technical Field

[0001] The present invention relates to the technical field of data fusion, and in particular to a method and system for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion. Background Art

[0002] Since the cold storage efficiency of the cold storage tank can not only intuitively understand the cold storage capacity of the cold storage tank during actual operation, but also evaluate the performance of the entire cold storage system, and can also provide data support for subsequent optimization of energy configuration and economic benefits, improvement of power grid stability and power generation efficiency, and promotion of energy structure optimization and sustainable development, it is crucial to effectively or accurately evaluate the cold storage efficiency of the cold storage tank.

[0003] In the prior art, generally, multiple sensors arranged at the inlet and outlet of the cold storage tank are first used to obtain multiple inlet temperature sequences and multiple outlet temperature sequences collected during the cold storage and release cycle. Then, the mean method is used to fuse all the inlet temperature sequences and all the outlet temperature sequences respectively to obtain the mean sequence of all the inlet temperature sequences and the mean sequence of all the outlet temperature sequences. After that, the cold storage efficiency corresponding to the cold storage and release cycle is calculated or evaluated through the mean sequence of all the inlet temperature sequences and the mean sequence of all the outlet temperature sequences obtained. However, since the working states of different sensors will be different, and the differences in the working states of different sensors will cause differences in the reference value or accuracy of the data collected by different sensors. This difference will lead to a lower reliability or reference value of the mean sequence obtained by the above mean method, and further lead to a lower effectiveness or accuracy of the calculated or evaluated cold storage efficiency. Therefore, how to improve the accuracy or effectiveness when evaluating the cold storage efficiency of the cold storage tank has become an urgent problem to be solved. Summary of the Invention

[0004] In order to solve the above problems, the present invention provides a method and system for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion. The specific technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion, including the following steps:

[0006] In the a-th cold storage and release cycle of the cold storage tank, obtain the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensors, where the monitoring sensors include all the inlet sensors and all the outlet sensors of the cold storage tank;

[0007] An evaluation index value of the monitoring sensor is obtained according to the difference between the temperature sequence to be analyzed of the monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, the correlation between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensor, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor;

[0008] According to the evaluation index values of all the imported sensors, the temperature sequences to be analyzed of all the imported sensors are weighted and fused to obtain an imported fusion temperature sequence; according to the evaluation index values of all the exported sensors, the temperature sequences to be analyzed of all the exported sensors are weighted and fused to obtain an exported fusion temperature sequence; according to the imported fusion temperature sequence and the exported fusion temperature sequence, the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle is obtained.

[0009] In a second aspect, an embodiment of the present invention provides a cold storage tank cold storage efficiency analysis system based on multi-source data fusion, including a memory and a processor, and the processor executes a computer program stored in the memory to implement the above-mentioned method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion.

[0010] Beneficial effects: First, in the a-th cold storage and release cycle of the cold storage tank, the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensor are obtained, and the monitoring sensors include all the imported sensors and all the exported sensors of the cold storage tank; then, according to the difference between the temperature sequence to be analyzed of the monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, the correlation between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensor, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor, the evaluation index value of the monitoring sensor is obtained; then, according to the evaluation index values of all the imported sensors, the temperature sequences to be analyzed of all the imported sensors are weighted and fused to obtain an imported fusion temperature sequence; according to the evaluation index values of all the exported sensors, the temperature sequences to be analyzed of all the exported sensors are weighted and fused to obtain an exported fusion temperature sequence; finally, according to the imported fusion temperature sequence and the exported fusion temperature sequence, the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle is obtained. And by weighting and fusing the temperature sequences to be analyzed of the sensors through the evaluation index values of the sensors, the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle obtained can be made more effective and accurate, that is, the present invention can improve the effectiveness and accuracy of the evaluation of the cold storage efficiency of the cold storage tank. Description of the Drawings

[0011] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0012] Figure 1 This is a flowchart of a method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to the present invention. Specific embodiments

[0013] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art belong to the scope protected by the embodiments of the present invention.

[0014] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0015] This embodiment provides a method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion, which is described in detail as follows:

[0016] As Figure 1 shown, the method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion includes the following steps:

[0017] Step S001, in the a-th cold storage and release cycle of the cold storage tank, obtain the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensors, where the monitoring sensors include all inlet sensors and all outlet sensors of the cold storage tank.

[0018] This embodiment mainly adaptively obtains the weights of the data collected by different inlet and outlet sensors when fusing by analyzing the performance or working state of different inlet and outlet sensors during the cold storage and release cycle, so as to improve the accuracy or effectiveness of evaluating the cold storage efficiency of the cold storage tank; in addition, for the convenience of analysis and understanding, in the following, the process of obtaining the cold storage efficiency of any cold storage tank in the a-th cold storage and release cycle will be used as an example for description or analysis.

[0019] In this embodiment, first, in the a-th charge-discharge cycle of the cold storage tank, the temperature sequence to be analyzed and the reference temperature sequence of each monitoring sensor of the cold storage tank are obtained. Here, a > 1, and the monitoring sensors in this embodiment refer to all inlet sensors and all outlet sensors of the cold storage tank. That is, in this embodiment, all inlet sensors and all outlet sensors of the cold storage tank are collectively referred to as monitoring sensors. An inlet sensor refers to a temperature sensor installed at the inlet position of the cold storage tank, and an outlet sensor refers to a temperature sensor installed at the outlet position of the cold storage tank. The number of inlet sensors and the number of outlet sensors in this embodiment are both greater than 1. And in this embodiment, the specific acquisition of the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensors is as follows:

[0020] For any monitoring sensor, the time series formed by all the temperature data monitored and collected by this monitoring sensor during the a-th charge-discharge cycle of the cold storage tank is recorded as the temperature sequence to be analyzed of this monitoring sensor, and the time series formed by all the temperature data monitored and collected by this monitoring sensor during the (a - 1)-th charge-discharge cycle of the cold storage tank is recorded as the reference temperature sequence of this monitoring sensor. And a charge-discharge cycle refers to the time unit for the ice storage system or the cold storage tank to complete the whole process of cold storage (storing cold energy) and cold release (releasing cold energy). Usually, a 24-hour period is a complete cycle.

[0021] In this embodiment, the acquisition frequencies of all monitoring sensors are not only the same but also synchronous. And if a certain moment is the temperature acquisition moment, then all monitoring sensors of the cold storage tank need to perform data acquisition at this moment, that is, all monitoring sensors of the cold storage tank can acquire a temperature data at this moment. In this embodiment, the acquisition frequencies of all monitoring sensors are set to empirical values. For example, if the cold storage tank in this embodiment is a natural stratified cold storage tank, then all monitoring sensors of this cold storage tank are generally set to perform data acquisition once every 5 to 10 seconds. In addition, when arranging sensors at the inlet and outlet of the cold storage tank, it is necessary to comply with the engineering technical specifications of the ice storage system. For example, if the cold storage tank in this embodiment is a closed pressure-bearing cold storage tank, then at least 2 groups of temperature sensors should be set for each inlet and outlet pipeline. The sensors need to be installed on a straight pipe section with a diameter of ≥ 10 times the pipe diameter, avoiding turbulence sources such as valves and elbows. And if the cold storage tank in this embodiment is a medium-sized closed pressure-bearing cold storage tank, generally 3 groups of temperature sensors are set in the inlet and outlet pipelines, and generally 2 to 3 sensors are included in one group of sensors.

[0022] Therefore, through the above process, this embodiment can obtain the temperature sequence to be analyzed and the reference temperature sequence of each monitoring sensor of the cold storage tank in the a-th charge-discharge cycle of the cold storage tank.

[0023] Step S002: Obtain the evaluation index value of the monitoring sensor based on the difference between the temperature sequence to be analyzed of the monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, the correlation between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensor, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor.

[0024] After obtaining the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensor, in this embodiment, the evaluation index value of each monitoring sensor is obtained based on the difference between the temperature sequence to be analyzed of each monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, the correlation between the temperature sequence to be analyzed of each monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensor, and the coefficient of variation of the temperature sequence to be analyzed of each monitoring sensor. The evaluation index value of the monitoring sensor can characterize the working state or working accuracy of the corresponding monitoring sensor during the a-th cold storage and release cycle. Then, in this embodiment, the specific process of obtaining the evaluation index value of the monitoring sensor needs to be described next. In addition, for the sake of understanding, in the following, the process of obtaining the evaluation index value of any monitoring sensor A will be taken as an example for description, that is, the process of obtaining the evaluation index value of monitoring sensor A is as follows:

[0025] First, obtain the first eigenvalue of monitoring sensor A during the a-th cold storage and release cycle based on the difference between the temperature sequence to be analyzed of monitoring sensor A and the reference temperature sequence of monitoring sensor A. And in this embodiment, the specific process of obtaining the first eigenvalue of monitoring sensor A during the a-th cold storage and release cycle is as follows:

[0026] First, denote the temperature sequence to be analyzed and the reference temperature sequence of monitoring sensor A as the first sequence and the second sequence respectively. Then, obtain all the temperature data belonging to the cold storage stage of the cold storage tank in the first sequence, and form a time series sequence with all the temperature data belonging to the cold storage stage of the cold storage tank obtained in the first sequence and denote it as the cold storage sequence to be analyzed. Obtain all the temperature data belonging to the cold release stage of the cold storage tank in the first sequence, and form a time series sequence with all the temperature data belonging to the cold release stage of the cold storage tank obtained in the first sequence and denote it as the cold release sequence to be analyzed. Then, obtain all the temperature data belonging to the cold storage stage of the cold storage tank in the second sequence, and form a time series sequence with all the temperature data belonging to the cold storage stage of the cold storage tank obtained in the second sequence and denote it as the reference cold storage sequence. Obtain all the temperature data belonging to the cold release stage of the cold storage tank in the second sequence, and form a time series sequence with all the temperature data belonging to the cold release stage of the cold storage tank obtained in the second sequence and denote it as the reference cold release sequence.

[0027] Next, obtain the temperature curve corresponding to the cold storage sequence to be analyzed, the temperature curve corresponding to the reference cold storage sequence, the temperature curve corresponding to the cold release sequence to be analyzed, and the temperature curve corresponding to the reference cold release sequence. Then, calculate the DTW distance between the temperature curve corresponding to the cold storage sequence to be analyzed and the temperature curve corresponding to the reference cold storage sequence, and denote it as the first DTW distance. Calculate the DTW distance between the temperature curve corresponding to the cold release sequence to be analyzed and the temperature curve corresponding to the reference cold release sequence, and denote it as the second DTW distance. After that, obtain the normalization value of the reciprocal of the result obtained by adding the preset hyperparameter, the first DTW distance, and the second DTW distance, and denote it as the first characterization value. Moreover, the calculation process of the DTW distance between any two sequences or two curves is a well-known technology. Next, continue to obtain the absolute value of the difference between the maximum value in the cold storage sequence to be analyzed and the maximum value in the reference cold storage sequence, and denote it as the first difference value. Obtain the absolute value of the difference between the minimum value in the cold storage sequence to be analyzed and the minimum value in the reference cold storage sequence, and denote it as the second difference value. Obtain the absolute value of the difference between the maximum value in the cold release sequence to be analyzed and the maximum value in the reference cold release sequence, and denote it as the third difference value. Obtain the absolute value of the difference between the minimum value in the cold release sequence to be analyzed and the minimum value in the reference cold release sequence, and denote it as the fourth difference value. Then, obtain the normalization value of the reciprocal of the result obtained by adding the preset hyperparameter, the first difference value, the second difference value, the third difference value, and the fourth difference value, and denote it as the second characterization value. Finally, perform weighted summation on the first characterization value and the second characterization value, and denote the result of the weighted summation as the first eigenvalue of the monitoring sensor in the a-th cold storage and release cycle.

[0028] In this embodiment, the process of obtaining the temperature curve is as follows: For the cold storage sequence to be analyzed, map each temperature data and the acquisition time of each temperature data in the cold storage sequence to be analyzed into a two-dimensional space to obtain the mapped data points of each temperature data in the cold storage sequence to be analyzed. Then, connect the mapped data points of each temperature data in the cold storage sequence to be analyzed in the order of the acquisition time, and denote the connected curve as the temperature curve corresponding to the cold storage sequence to be analyzed. The horizontal axis of the two-dimensional space represents time, and the vertical axis represents temperature data. The abscissa value of the mapped data point of the temperature data is the acquisition time of the corresponding temperature data, and the ordinate value is the corresponding temperature data. Moreover, the methods for obtaining the temperature curve corresponding to the reference cold storage sequence, the temperature curve corresponding to the cold release sequence to be analyzed, and the temperature curve corresponding to the reference cold release sequence are the same as the method for obtaining the temperature curve corresponding to the cold storage sequence to be analyzed described above, so they will not be described in detail here.

[0029] In addition, the calculation expression of the first eigenvalue of the monitoring sensor A in the a-th cold storage and release cycle is:

[0030] To monitor the first eigenvalue of the monitoring sensor A in the a-th cold storage and release cycle, Norm() is a normalization function, is the first DTW distance, is the second DTW distance, D1 is the first difference value, D2 is the second difference value, D3 is the third difference value, D4 is the fourth difference value, w is a preset hyperparameter. To prevent the denominator from being 0, in this embodiment, the preset hyperparameter is set to 1.

[0031] And when 、 、D1, D2, D3, and D4 are smaller, the value of is larger, and the larger the value of, the higher the similarity between the temperature sequence to be analyzed of the monitoring sensor A and the reference temperature sequence. The higher the similarity, the smaller the probability that the monitoring sensor A has a working abnormality or the smaller the possibility that the monitoring sensor A is interfered during the a-th cold storage and release cycle. Then it indicates that the working state or stability of the monitoring sensor A during the a-th cold storage and release cycle is better or the reliability of the data collected by the monitoring sensor A during the a-th cold storage and release cycle is higher; on the contrary, when the value of is smaller, the lower the similarity between the temperature sequence to be analyzed of the monitoring sensor A and the reference temperature sequence. The lower the similarity, the greater the probability that the monitoring sensor A has a working abnormality or the greater the possibility that the monitoring sensor A is interfered during the a-th cold storage and release cycle. Then it indicates that the working state or stability of the monitoring sensor A during the a-th cold storage and release cycle is worse or the reliability of the data collected by the monitoring sensor A during the a-th cold storage and release cycle is lower; in addition, the reference temperature sequence is the data corresponding to the (a - 1)-th cold storage and release cycle. Therefore, by comparing the data differences collected by the same sensor in adjacent cold storage and release cycles, the probability of a working abnormality or the possibility of being interfered by the corresponding sensor can be reflected.

[0032] Then, according to the correlation between the temperature sequence to be analyzed of the monitoring sensor A and the temperature sequences to be analyzed of other monitoring sensors of the same type except the monitoring sensor A, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor A, the second eigenvalue of the monitoring sensor A in the a-th cold storage and release cycle is obtained; and in this embodiment, the specific process of obtaining the second eigenvalue of the monitoring sensor A in the a-th cold storage and release cycle is as follows:

[0033] First, obtain the coefficient of variation corresponding to the temperature sequence to be analyzed of the monitoring sensor A. The coefficient of variation is the ratio of the standard deviation of the temperature sequence to be analyzed of the monitoring sensor A to the mean of the temperature sequence to be analyzed of the monitoring sensor A. The coefficient of variation can reflect the data stability. The smaller the coefficient of variation, the more stable the data in the sequence. And the more stable the data change in the sequence, the smaller the probability that the monitoring sensor A has a working anomaly or the smaller the possibility that the monitoring sensor A is interfered, and it also indicates that the reliability of the data collected by the monitoring sensor A in the a-th cold storage and release cycle is higher. Then, record the normalized value of the reciprocal of the result obtained by adding the preset hyperparameter and the coefficient of variation corresponding to the temperature sequence to be analyzed of the monitoring sensor A as the first index value of the monitoring sensor A.

[0034] After that, among all the monitoring sensors of the cold storage tank, obtain all the monitoring sensors that are of the same type as the monitoring sensor A except the monitoring sensor A. And record the set composed of all the monitoring sensors that are of the same type as the monitoring sensor A except the monitoring sensor A as the subset of the monitoring sensor A. Record all the monitoring sensors in the subset as comparison sensors. And if the monitoring sensor A is an imported sensor and another monitoring sensor is also an imported sensor, then these two monitoring sensors are of the same type of monitoring sensors. If the monitoring sensor A is an imported sensor and another monitoring sensor is an exported sensor, then these two monitoring sensors are not of the same type of monitoring sensors. Then, obtain the Spearman correlation coefficient between the temperature sequence to be analyzed of the obtained monitoring sensor A and the temperature sequence to be analyzed of each comparison sensor in the subset, and record it as the correlation coefficient of the corresponding comparison sensor. The larger the Spearman correlation coefficient, the greater the degree of correlation. And the greater the degree of correlation, the smaller the probability that the monitoring sensor A has a working anomaly or the smaller the possibility that the monitoring sensor A is interfered, and it also indicates that the reliability of the data collected by the monitoring sensor A in the a-th cold storage and release cycle is higher. After that, obtain the sum of the correlation coefficient of each comparison sensor and the preset first constant, and record it as the initial correlation characterization value of the corresponding comparison sensor. Immediately afterwards, obtain the reference value characterization value of each comparison sensor in the subset, and record the product of the reference value characterization value of each comparison sensor and the initial correlation characterization value of the corresponding comparison sensor as the target correlation characterization value of the corresponding comparison sensor. Then, obtain the normalized value of the accumulated result of the target correlation characterization values of all the comparison sensors in the subset, and record it as the second index value of the monitoring sensor A. Finally, perform a weighted sum of the first index value and the second index value of the monitoring sensor A, and record the weighted sum result as the second characteristic value of the monitoring sensor A in the a-th cold storage and release cycle.

[0035] In this embodiment, the process of obtaining the reference value characterization value of each comparison sensor in the subset is as follows:

[0036] For any comparison sensor B in the subset, first obtain the weighted sum value of the first index value and the first characteristic value of the comparison sensor B, and denote it as the value to be processed of the comparison sensor B. The methods for obtaining the first index value and the first characteristic value of the comparison sensor B are the same as those for obtaining the first index value and the first characteristic value of the monitoring sensor A. And the calculation expression for the value to be processed of the comparison sensor B is , where is the first characteristic value of the comparison sensor B, is the first index value of the comparison sensor B.

[0037] Then obtain the sum of the values to be processed of all the comparison sensors in the subset, and use it as the comprehensive value to be processed. Take the ratio of the value to be processed of each comparison sensor in the subset to the comprehensive value to be processed as the reference value representation value of the corresponding comparison sensor; and the reference value representation value of the comparison sensor is mainly used to measure the participation degree of the initial correlation representation values of each comparison sensor when calculating the second index value of the monitoring sensor A. The larger the reference value representation value of the comparison sensor, the better the working state or stability of the corresponding comparison sensor in the a-th cold storage and release cycle, or the higher the reliability of the data collected by the corresponding comparison sensor in the a-th cold storage and release cycle. Then, when calculating the second index value of the monitoring sensor A, the participation degree of the initial correlation representation values of these comparison sensors should be made larger.

[0038] In addition, the calculation expression for the second characteristic value of the monitoring sensor A in the a-th cold storage and release cycle is:

[0039] where is the second characteristic value of the monitoring sensor A in the a-th cold storage and release cycle, V is the coefficient of variation corresponding to the temperature sequence to be analyzed of the monitoring sensor A, J is the number of comparison sensors in the subset of the monitoring sensor A, is the reference value representation value of the j-th comparison sensor in the subset of the monitoring sensor A, is the Spearman correlation coefficient between the temperature sequence to be analyzed of the monitoring sensor A and the temperature sequence to be analyzed of the j-th comparison sensor in the subset, and it is also the correlation coefficient of the j-th comparison sensor. s1 is a preset first constant, and w is a preset hyperparameter; in this embodiment, the value of s1 is set to 1, and the purpose is to make the initial correlation representation value of the comparison sensor non-negative, while is to make fall within the range of 0 to 1; the preset hyperparameter here is also to prevent the denominator from being 0.

[0040] And in this embodiment, when V is smaller and is larger, it indicates that has a larger value. And when has a larger value, it more indicates that the correlation between the monitoring sensor A and the comparison sensors in its corresponding subset is higher, the data change in the temperature sequence to be analyzed of the monitoring sensor A is more stable, and it also indicates that in the a-th cold storage and release cycle, the probability of the monitoring sensor A having an abnormal operation is smaller or the possibility of the monitoring sensor A being interfered is smaller, or it indicates that the working state or stability of the monitoring sensor A in the a-th cold storage and release cycle is better, and the reliability of the data collected by the monitoring sensor A in the a-th cold storage and release cycle is higher; conversely, when V is larger and is smaller, it indicates that has a smaller value. And when has a smaller value, it more indicates that the correlation between the monitoring sensor A and the comparison sensors in its corresponding subset is lower, the data change in the temperature sequence to be analyzed of the monitoring sensor A is more unstable, and it also indicates that in the a-th cold storage and release cycle, the probability of the monitoring sensor A having an abnormal operation is larger or the possibility of the monitoring sensor A being interfered is larger, or it indicates that the working state or stability of the monitoring sensor A in the a-th cold storage and release cycle is worse, and the reliability of the data collected by the monitoring sensor A in the a-th cold storage and release cycle is lower.

[0041] After that, the first eigenvalue of the monitoring sensor A in the a-th cold storage and release cycle and the second eigenvalue of the monitoring sensor A in the a-th cold storage and release cycle are weighted and summed, and the result of the weighted sum is used as the evaluation index value of the monitoring sensor A, that is, the evaluation index value of the monitoring sensor A is Moreover, the larger the evaluation index value of the monitoring sensor A is, the better the working state of the monitoring sensor A in the a-th cold storage and release cycle is, or in other words, the smaller the interference of other external factors on the monitoring sensor A is, and the better the performance of the monitoring sensor A itself. The other external factors can refer to the water flow rate, which also indicates that the reliability of the data collected by the monitoring sensor A in the a-th cold storage and release cycle is higher. Then, when calculating the cold storage efficiency of the a-th cold storage and release cycle, the reference value or participation degree of the temperature sequence to be analyzed of the monitoring sensor A should be greater. That is, when the evaluation index value of the monitoring sensor A is larger, when subsequently fusing the temperature sequences to be analyzed of all monitoring sensors of the same type as the monitoring sensor A, the fusion weight of the temperature sequence to be analyzed of the monitoring sensor A should be greater; conversely, when the evaluation index value of the monitoring sensor A is smaller, it indicates that the working state of the monitoring sensor A in the a-th cold storage and release cycle is worse, or in other words, the interference of other external factors on the monitoring sensor A is larger, and the performance of the monitoring sensor A itself is poorer, which also indicates that the reliability of the data collected by the monitoring sensor A in the a-th cold storage and release cycle is lower. Then, when calculating the cold storage efficiency of the a-th cold storage and release cycle, the reference value or participation degree of the temperature sequence to be analyzed of the monitoring sensor A should be smaller. That is, when the evaluation index value of the monitoring sensor A is smaller, when subsequently fusing the temperature sequences to be analyzed of all monitoring sensors of the same type as the monitoring sensor A, the fusion weight of the temperature sequence to be analyzed of the monitoring sensor A should be smaller.

[0042] Step S003: Based on the evaluation index values of all inlet sensors, perform weighted fusion on the temperature sequences to be analyzed of all inlet sensors to obtain an inlet fusion temperature sequence; based on the evaluation index values of all outlet sensors, perform weighted fusion on the temperature sequences to be analyzed of all outlet sensors to obtain an outlet fusion temperature sequence; based on the inlet fusion temperature sequence and the outlet fusion temperature sequence, obtain the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle.

[0043] In the following, this embodiment will perform weighted fusion on the temperature sequences to be analyzed of all inlet sensors of the cold storage tank based on the evaluation index values of all inlet sensors of the cold storage tank to obtain an inlet fusion temperature sequence, and perform weighted fusion on the temperature sequences to be analyzed of all outlet sensors of the cold storage tank based on the evaluation index values of all outlet sensors of the cold storage tank to obtain an outlet fusion temperature sequence. Subsequently, the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle will be evaluated based on the inlet fusion temperature sequence and the outlet fusion temperature sequence. Then, the specific processes of the inlet fusion temperature sequence and the outlet fusion temperature sequence in this embodiment are as follows:

[0044] First, obtain the sum of the evaluation index values of all inlet sensors of the cold storage tank, and denote it as the first comprehensive index value. Then, denote the ratio of the evaluation index value of each inlet sensor of the cold storage tank to the first comprehensive index value as the fusion weight value of the corresponding inlet sensor in the a-th charging and discharging cycle. After that, take the product of the fusion weight value of each inlet sensor in the a-th charging and discharging cycle and the temperature sequence to be analyzed of the corresponding inlet sensor as the weighted inlet temperature sequence of the corresponding inlet sensor. Finally, denote the sequence obtained by adding up the weighted inlet temperature sequences of all inlet sensors of the cold storage tank as the inlet fusion temperature sequence, and the b-th data in the inlet fusion temperature sequence is the sum of the b-th data in all weighted inlet temperature sequences. The h-th data in the weighted inlet temperature sequence of any inlet sensor is the result of multiplying the h-th data in the temperature sequence to be analyzed of the inlet sensor by the fusion weight value of the inlet sensor in the a-th charging and discharging cycle. Immediately afterwards, obtain the sum of the evaluation index values of all outlet sensors of the cold storage tank, and denote it as the second comprehensive index value. Then, denote the ratio of the evaluation index value of each outlet sensor of the cold storage tank to the second comprehensive index value as the fusion weight value of the corresponding outlet sensor in the a-th charging and discharging cycle. After that, take the product of the fusion weight value of each outlet sensor in the a-th charging and discharging cycle and the temperature sequence to be analyzed of the corresponding outlet sensor as the weighted outlet temperature sequence of the corresponding outlet sensor. Finally, denote the sequence obtained by adding up the weighted inlet temperature sequences of all outlet sensors of the cold storage tank as the outlet fusion temperature sequence, and the f-th data in the outlet fusion temperature sequence is the sum of the f-th data in all weighted outlet temperature sequences. The r-th data in the weighted outlet temperature sequence of any outlet sensor is the result of multiplying the r-th data in the temperature sequence to be analyzed of the outlet sensor by the fusion weight value of the outlet sensor in the a-th charging and discharging cycle. And in this embodiment, it is required that the data in the inlet fusion temperature sequence and the outlet fusion temperature sequence be retained to one decimal place.

[0045] After obtaining the inlet fusion temperature sequence and the outlet fusion temperature sequence, the cold storage efficiency of the cold storage tank in the a-th charging and discharging cycle is calculated by a well-known calculation method for the cold storage efficiency of the cold storage tank. However, in this embodiment, the temperature data used in calculating the cold storage efficiency of the cold storage tank in the a-th charging and discharging cycle all come from the inlet fusion temperature sequence and the outlet fusion temperature sequence. For example, the commonly used calculation formula for calculating the cold storage efficiency of the cold storage tank in the a-th charging and discharging cycle is , where is the cold storage efficiency of the cold storage tank in the a-th charging and discharging cycle, is the total cold release amount in the a-th charging and discharging cycle, is the total cold storage amount in the a-th charging and discharging cycle, is the cold release time in the a-th charging and discharging cycle, is the cold storage time in the a-th cold storage and release cycle, is the mass flow rate during cold release in the a-th cold storage and release cycle, is the mass flow rate during cold storage in the a-th cold storage and release cycle, c is the specific heat capacity of the cold carrier medium, generally the specific heat capacity of water, T1 is the temperature at the inlet of the cold storage tank during the cold release stage in the a-th cold storage and release cycle, T2 is the temperature at the outlet of the cold storage tank during the cold release stage in the a-th cold storage and release cycle, T3 is the temperature at the outlet of the cold storage tank during the cold storage stage in the a-th cold storage and release cycle, T4 is the temperature at the inlet of the cold storage tank during the cold storage stage in the a-th cold storage and release cycle, and in this embodiment, T1 and T4 are data in the inlet fusion temperature sequence, T2 and T3 are data in the outlet fusion temperature sequence. The mass flow rate during cold release refers to the mass of the cold carrier medium flowing out of the cold storage tank and used for cooling per unit time, and the mass flow rate during cold storage refers to the mass of the cold carrier medium flowing into the cold storage tank and storing cold per unit time.

[0046] Thus, through the above process, the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle is obtained in this embodiment, and based on the data obtained by the fusion method provided in this embodiment, the effectiveness and accuracy of the cold storage efficiency evaluation of the cold storage tank can be improved.

[0047] A cold storage tank cold storage efficiency analysis system based on multi-source data fusion in this embodiment includes a memory and a processor. The processor executes the computer program stored in the memory to implement the above-mentioned cold storage tank cold storage efficiency analysis method based on multi-source data fusion.

[0048] In summary, in this embodiment, in the a-th charging and discharging cycle of the cold storage tank, the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensors are obtained. The monitoring sensors include all inlet sensors and all outlet sensors of the cold storage tank. Then, based on the difference between the temperature sequence to be analyzed of the monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, the correlation between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensor, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor, the evaluation index value of the monitoring sensor is obtained. Then, based on the evaluation index values of all inlet sensors, the temperature sequences to be analyzed of all inlet sensors are weighted and fused to obtain the inlet fused temperature sequence. Based on the evaluation index values of all outlet sensors, the temperature sequences to be analyzed of all outlet sensors are weighted and fused to obtain the outlet fused temperature sequence. Finally, based on the inlet fused temperature sequence and the outlet fused temperature sequence, the cold storage efficiency of the cold storage tank in the a-th charging and discharging cycle is obtained. Moreover, the method of weighting and fusing the temperature sequences to be analyzed of the sensors according to the evaluation index values of the sensors in this embodiment can make the cold storage efficiency of the cold storage tank in the a-th charging and discharging cycle more effective and accurate, that is, this embodiment can improve the effectiveness and accuracy of the evaluation of the cold storage efficiency of the cold storage tank.

[0049] The above embodiments are only used to illustrate the technical solutions of the present application, not to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion, characterized in that, The method includes the following steps: In the a-th cold storage and release cycle of the cold storage tank, obtain the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensors, where the monitoring sensors include all inlet sensors and all outlet sensors of the cold storage tank; Based on the difference between the temperature sequence to be analyzed of the monitoring sensors and the reference temperature sequence of the corresponding monitoring sensors, the correlation between the temperature sequence to be analyzed of the monitoring sensors and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensors, and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensors, obtain the evaluation index value of the monitoring sensors; Based on the evaluation index values of all inlet sensors, determine the fusion weight value of each inlet sensor in the a-th cold storage and release cycle, and perform weighted fusion on the temperature sequences to be analyzed of all inlet sensors according to the fusion weight value to obtain the inlet fusion temperature sequence; based on the evaluation index values of all outlet sensors, determine the fusion weight value of each outlet sensor in the a-th cold storage and release cycle, and perform weighted fusion on the temperature sequences to be analyzed of all outlet sensors according to the fusion weight value to obtain the outlet fusion temperature sequence; based on the inlet fusion temperature sequence and the outlet fusion temperature sequence, obtain the cold storage efficiency of the cold storage tank in the a-th cold storage and release cycle.

2. The method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to claim 1, characterized in that, The method for obtaining the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensors includes: For any monitoring sensor, record the temperature sequence monitored by the monitoring sensor in the a-th cold storage and release cycle of the cold storage tank as the temperature sequence to be analyzed of the monitoring sensor, and record the temperature sequence monitored by the monitoring sensor in the (a - 1)-th cold storage and release cycle of the cold storage tank as the reference temperature sequence of the monitoring sensor, where a > 1.

3. The cold storage efficiency analysis method of the cold storage tank based on multi-source data fusion according to claim 1, wherein The method for obtaining the evaluation index value of the monitoring sensors includes: For any monitoring sensor, based on the difference between the temperature sequence to be analyzed of the monitoring sensor and the reference temperature sequence of the corresponding monitoring sensor, obtain the first characteristic value of the monitoring sensor in the a-th cold storage and release cycle, and based on the correlation between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequences to be analyzed of other monitoring sensors of the same type except the corresponding monitoring sensors and the coefficient of variation of the temperature sequence to be analyzed of the monitoring sensor, obtain the second characteristic value of the monitoring sensor in the a-th cold storage and release cycle. Take the result of weighted summation of the first characteristic value and the second characteristic value as the evaluation index value of the monitoring sensor. All inlet sensors of the cold storage tank belong to the same type of monitoring sensors, and all outlet sensors of the cold storage tank belong to the same type of monitoring sensors.

4. The method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to claim 3, wherein The method for obtaining the first characteristic value of the monitoring sensor in the a-th cold storage and release cycle includes: Denote the temperature sequence to be analyzed and the reference temperature sequence of the monitoring sensor as the first sequence and the second sequence respectively; Denote the sequence formed by all the data in the first sequence that belongs to the cold storage stage of the cold storage tank as the cold storage sequence to be analyzed, denote the sequence formed by all the data in the first sequence that belongs to the cold release stage of the cold storage tank as the cold release sequence to be analyzed, denote the sequence formed by all the data in the second sequence that belongs to the cold storage stage of the cold storage tank as the reference cold storage sequence, and denote the sequence formed by all the data in the second sequence that belongs to the cold release stage of the cold storage tank as the reference cold release sequence; Obtain the temperature curves corresponding to the cold storage sequence to be analyzed, the reference cold storage sequence, the cold release sequence to be analyzed, and the reference cold release sequence. Denote the DTW distance between the temperature curve corresponding to the cold storage sequence to be analyzed and the temperature curve corresponding to the reference cold storage sequence as the first DTW distance, denote the DTW distance between the temperature curve corresponding to the cold release sequence to be analyzed and the temperature curve corresponding to the reference cold release sequence as the second DTW distance, and denote the normalized value of the reciprocal of the result obtained by adding the preset hyperparameter, the first DTW distance, and the second DTW distance as the first characterization value; Denote the absolute value of the difference between the maximum value in the cold storage sequence to be analyzed and the maximum value in the reference cold storage sequence as the first difference value, denote the absolute value of the difference between the minimum value in the cold storage sequence to be analyzed and the minimum value in the reference cold storage sequence as the second difference value, denote the absolute value of the difference between the maximum value in the cold release sequence to be analyzed and the maximum value in the reference cold release sequence as the third difference value, denote the absolute value of the difference between the minimum value in the cold release sequence to be analyzed and the minimum value in the reference cold release sequence as the fourth difference value, and denote the normalized value of the reciprocal of the result obtained by adding the preset hyperparameter, the first difference value, the second difference value, the third difference value, and the fourth difference value as the second characterization value; Denote the result of weighted summation of the first characterization value and the second characterization value as the first eigenvalue of the monitoring sensor in the a-th cold storage and release cycle.

5. The method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to claim 3, wherein The method for obtaining the second eigenvalue of the monitoring sensor in the a-th cold storage and release cycle includes: Denote the normalized value of the reciprocal of the result obtained by adding the preset hyperparameter and the coefficient of variation corresponding to the temperature sequence to be analyzed of the monitoring sensor as the first index value of the monitoring sensor; Among all the monitoring sensors, obtain the set consisting of all the monitoring sensors that are of the same type as the monitoring sensor but excluding the monitoring sensor itself, and denote it as the subset. Denote all the monitoring sensors in the subset as comparison sensors; obtain the Spearman correlation coefficient between the temperature sequence to be analyzed of the monitoring sensor and the temperature sequence to be analyzed of each comparison sensor, and denote it as the correlation coefficient of the corresponding comparison sensor. Denote the sum of the correlation coefficient of the comparison sensor and a preset first constant as the initial correlation characterization value of the corresponding comparison sensor; obtain the reference value characterization value of the comparison sensor, and denote the product of the initial correlation characterization value of the comparison sensor and the reference value characterization value of the corresponding comparison sensor as the target correlation characterization value of the corresponding comparison sensor; denote the normalized value of the result obtained by accumulating the target correlation characterization values of all the comparison sensors in the subset as the second index value of the monitoring sensor. Denote the result of weighted summation of the first index value and the second index value of the monitoring sensor as the second characteristic value of the monitoring sensor in the a-th cold storage and release cycle.

6. The method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to claim 5, wherein The method for obtaining the reference value characterization value of the comparison sensor includes: For any comparison sensor, obtain the result of weighted summation of the first index value of the comparison sensor and the first characteristic value of the comparison sensor, and denote it as the value to be processed of the comparison sensor. Take the sum of the values to be processed of all the comparison sensors in the subset as the comprehensive value to be processed, and take the ratio of the value to be processed of each comparison sensor in the subset to the comprehensive value to be processed as the reference value characterization value of the corresponding comparison sensor.

7. The method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to claim 1, characterized in that, The method for obtaining the inlet fusion temperature sequence includes: Obtain the sum of the evaluation index values of all the inlet sensors of the cold storage tank, and denote it as the first comprehensive index value; denote the ratio of the evaluation index value of each inlet sensor of the cold storage tank to the first comprehensive index value as the fusion weight value of the corresponding inlet sensor in the a-th cold storage and release cycle; take the product of the fusion weight value of each inlet sensor in the a-th cold storage and release cycle and the temperature sequence to be analyzed of the corresponding inlet sensor as the weighted inlet temperature sequence of the corresponding inlet sensor; denote the sequence obtained by adding up the weighted inlet temperature sequences of all the inlet sensors of the cold storage tank as the inlet fusion temperature sequence.

8. The method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion according to claim 1, wherein The method for obtaining the outlet fusion temperature sequence includes: Obtain the sum of the evaluation index values of all the outlet sensors of the cold storage tank, and denote it as the second comprehensive index value; denote the ratio of the evaluation index value of each outlet sensor of the cold storage tank to the second comprehensive index value as the fusion weight value of the corresponding outlet sensor in the a-th cold storage and release cycle; take the product of the fusion weight value of each outlet sensor in the a-th cold storage and release cycle and the temperature sequence to be analyzed of the corresponding outlet sensor as the weighted outlet temperature sequence of the corresponding outlet sensor; denote the sequence obtained by adding up the weighted outlet temperature sequences of all the outlet sensors of the cold storage tank as the outlet fusion temperature sequence.

9. A cold storage efficiency analysis system for cold storage tanks based on multi-source data fusion, comprising a memory and a processor, characterized in that, The processor executes the computer program stored in the memory to implement a method for analyzing the cold storage efficiency of a cold storage tank based on multi-source data fusion as described in any one of claims 1-8.

Images