Data optimization method and system for ultrasonic flow meter

Abstract

The application provides a data optimization method and system of an ultrasonic flowmeter, feature contents of to-be-analyzed ultrasonic data are extracted, and a bias degree is obtained; attribute drawing data corresponding to each feature content is obtained, each feature content is determined with a first common factor of a to-be-selected type, then the bias degree of each feature content and each first common factor are used to determine a second common factor of each to-be-selected type corresponding to the to-be-analyzed ultrasonic data, and then the type to which the to-be-analyzed ultrasonic data belongs is determined from each to-be-selected type through each second common factor. The application optimizes the ultrasonic data before metering and counting of the ultrasonic flowmeter, can analyze the feature contents of the to-be-analyzed ultrasonic data in multiple dimensions on the premise that the type to which the ultrasonic data belongs is unknown, can accurately determine the type of the ultrasonic data, and thus can more accurately optimize the ultrasonic data and improve the metering efficiency of the ultrasonic flowmeter.

Description

Technical Field

[0001] This application relates to the field of ultrasonic data optimization technology, and more specifically, to a data optimization method and system for an ultrasonic flow meter. Background Technology

[0002] An ultrasonic flow meter measures flow rate by detecting the effect of fluid flow on an ultrasonic beam (or ultrasonic pulse). It typically employs advanced multi-pulse technology, digital signal processing, and error correction techniques, making it more adaptable to industrial environments and thus widely applicable in fields such as petroleum, chemical, metallurgy, power, water supply and drainage, and gas.

[0003] Technical personnel discovered that ultrasonic flow meters suffer from low data measurement accuracy during use. Therefore, repeated manual optimization of each measurement is required to obtain accurate results. This not only wastes time and labor costs but also reduces the company's efficiency.

[0004] Through long-term research, relevant technical personnel have discovered that the main factors affecting the measurement accuracy of ultrasonic flow meters include the data acquisition method and the software's calculation method. However, improving the measurement accuracy of ultrasonic flow meters from these two aspects remains a difficult technical problem to solve at present. Summary of the Invention

[0005] To address the technical problems existing in related technologies, this application provides a data optimization method and system for ultrasonic flow meters.

[0006] Firstly, a data optimization method for an ultrasonic flow meter is provided. The method includes: extracting feature content from ultrasonic data to be analyzed and determining the bias of each feature content, wherein the feature content is a feature in the ultrasonic data to be analyzed that represents the core content of the ultrasonic data, and the bias indicates the importance of the corresponding feature content to the ultrasonic data; obtaining attribute characterization data corresponding to each feature content, wherein the attribute characterization data is data used to assist in understanding the description of the feature content; and determining a first common factor between each feature content and each candidate category through the attribute characterization data and category description data of each candidate category. The category description data is used to characterize the characteristics of the candidate categories. The first common factor is an evaluation value used to assess the degree of correlation between the feature content and the candidate categories. Through the bias of each feature content and each of the first common factors, a second common factor is determined between the ultrasound data to be analyzed and each of the candidate categories. The second common factor is an evaluation value used to assess the degree of correlation between the ultrasound data to be analyzed and the candidate categories. Through each of the second common factors, the category to which the ultrasound data to be analyzed belongs is determined from each of the candidate categories, and the category to which the ultrasound data to be analyzed belongs is optimized to determine the optimization result of the ultrasound data category.

[0007] In one independently implemented embodiment, the extraction of feature content from the ultrasound data to be analyzed includes: performing local feature processing on the ultrasound data to be analyzed to obtain X first local features of the ultrasound data to be analyzed; deleting first local features belonging to a target filtering feature set from each of the first local features to determine Y second local features; the second local features include the remaining first local features after deletion; wherein X is greater than 0 and X is greater than or equal to Y; obtaining the feature content of the ultrasound data to be analyzed through each of the second local features; wherein the target filtering feature set includes a filtering feature set corresponding to a target initial data source, and the target initial data source includes the initial data source to which the ultrasound data to be analyzed belongs.

[0008] In one standalone embodiment, the method for constructing the target filtering feature set includes: performing local feature processing on each ultrasonic data belonging to the target initial data source to obtain several third local features; sequentially determining a first proportion corresponding to each of the third local features; the first proportion corresponding to the third local feature refers to the proportion of the ultrasonic dataset in the target initial data source that covers the third local feature to the total number of ultrasonic data in the target initial data source; and constructing the target filtering feature set by combining the third local features whose first proportion is greater than a specified value.

[0009] In one independently implemented embodiment, determining the first proportion specification value includes: determining a fourth local feature from each of the third local features based on the real-time proportion specification value; the fourth local feature includes third local features whose first proportion is not less than the real-time proportion specification value; determining the number of remaining features corresponding to each ultrasonic data belonging to the target initial data source, the number of remaining features corresponding to the ultrasonic data being the number of third local features remaining after deleting the fourth local feature from each of the third local features of the ultrasonic data; determining a second proportion of the ultrasonic dataset whose number of remaining features is not less than the feature number specification value to the total number of ultrasonic data belonging to the target initial data source; when the second proportion is not greater than the second proportion specification value, determining the real-time proportion specification value as the first proportion specification value; when the second proportion is greater than the second proportion specification value, adjusting the real-time proportion specification value according to the adjustment value, and returning to the step of determining the fourth local feature from each of the third local features based on the real-time proportion specification value.

[0010] In one independently implemented embodiment, obtaining the feature content of the ultrasound data to be analyzed through each of the second local features includes: distributing the second local features to obtain fifth local features; each of the fifth local features covers at least two consecutive second local features; determining a sixth local feature from each of the fifth local features; the sixth local feature includes key features belonging to the fifth local features; determining a seventh local feature from each of the sixth local features; the seventh local feature does not cover each of the sixth local features; and combining the seventh local feature to obtain the feature content of the ultrasound data to be analyzed.

[0011] In one independently implemented embodiment, obtaining the attribute characterization data corresponding to each of the aforementioned feature contents includes: obtaining real-time ultrasonic mapping content corresponding to each of the aforementioned feature contents, wherein the real-time ultrasonic mapping content corresponding to the feature content is obtained by matching the feature content through a real-time data matching unit; and obtaining the attribute characterization data corresponding to each of the aforementioned feature contents through the real-time ultrasonic mapping content corresponding to each of the aforementioned feature contents.

[0012] In one independently implemented embodiment, obtaining the real-time ultrasound mapping content corresponding to each of the aforementioned feature contents includes: searching for candidate feature contents corresponding to each of the aforementioned feature contents in a database; the database records the association between candidate feature contents and candidate matching information, wherein the candidate matching information is obtained by matching the corresponding candidate feature contents through the real-time data matching unit; when a candidate feature content corresponding to a feature content is found, obtaining the real-time ultrasound mapping content corresponding to that feature content based on the candidate matching information matched by the found candidate feature content; when no candidate feature content corresponding to a feature content is found, calling the real-time data matching unit to match that feature content to obtain the real-time ultrasound mapping content corresponding to that feature content.

[0013] In one independently implemented embodiment, the method further includes: obtaining importance level parameters for each of the candidate categories; determining the category to which the ultrasonic data to be analyzed belongs from each of the candidate categories using each of the second common factors, and optimizing the category to which the ultrasonic data to be analyzed belongs to determine the optimization result of the ultrasonic data category, including: determining the fifth common factor between the ultrasonic data to be analyzed and each of the candidate categories using each of the second common factors and the importance level parameters for each of the candidate categories; determining the category to which the ultrasonic data to be analyzed belongs from each of the candidate categories using each of the fifth common factors, and optimizing the category to which the ultrasonic data to be analyzed belongs to determine the optimization result of the ultrasonic data category.

[0014] In one independently implemented embodiment, determining the first common factor between each feature content and each candidate species using the attribute characterization data and the species description data of each candidate species includes: determining the third common factor between each attribute characterization data and each candidate species using the attribute characterization data and the species labels of each candidate species; and determining the first common factor between each feature content and each candidate species using the third common factor.

[0015] In one independently implemented embodiment, determining the third common factor between each attribute characterization data and each candidate species using the attribute characterization data and the species labels of each candidate species includes: determining the intersection features between each attribute characterization data and the species labels of each candidate species using the attribute characterization data and the species labels of each candidate species; determining the target intersection features between each attribute characterization data and the species labels of each candidate species from the intersection features; and determining the total features of the target intersection features between each attribute characterization data and the species labels of each candidate species. The third proportion of the total number of features of the category labels of each candidate category is used to determine the third common factor between each attribute characterization data and the category labels of each candidate category. This factor is determined by the third proportion, the first feature point of the target intersection feature between each attribute characterization data and the category labels of each candidate category in the corresponding attribute characterization data, and the first abnormal text probability of the target intersection feature between each attribute characterization data and the category labels of each candidate category. The target intersection feature between the attribute characterization data and the category labels of the candidate category is not included in the intersection features of the attribute characterization data and the category labels of the candidate category other than itself.

[0016] In one independently implemented embodiment, the method further includes: determining a fourth common factor between each attribute characterization data and each candidate species using the attribute characterization data, the pre-set species association features of each candidate species, and the pre-set species association features of each candidate species and the pre-set related parameters of the corresponding candidate species; the step of determining a first common factor between each feature content and each candidate species using the third common factor includes: determining a first common factor between each feature content and each candidate species using the third common factor and the fourth common factor.

[0017] In one independently implemented embodiment, determining the fourth common factor between each attribute characterization data and each candidate type through the attribute characterization data, the pre-set type association features of each candidate type, and the pre-set type association features of each candidate type and the pre-set related parameters of the corresponding candidate type includes: determining the fourth common factor between each attribute characterization data and each candidate type through the second feature points of the pre-set type association features of each candidate type in the attribute characterization data, the second abnormal text probability of the pre-set type association features of each candidate type, and the pre-set related parameters of the pre-set type association features of each candidate type and the corresponding candidate type.

[0018] Secondly, a data optimization system for an ultrasonic flow meter is provided, comprising a processor and a memory that communicate with each other, wherein the processor is used to read a computer program from the memory and execute it to implement the method described above.

[0019] This application provides a data optimization method and system for ultrasonic flow meters. The method extracts feature content from the ultrasonic data to be analyzed and obtains the bias degree of each feature content. Then, it obtains attribute characterization data corresponding to each feature content. Using the attribute characterization data, it determines the first common factor between each feature content and the candidate categories. Then, using the bias degree of each feature content and the first common factor, it determines the second common factor between the ultrasonic data to be analyzed and each candidate category. Finally, using the second common factor, it determines the category to which the ultrasonic data to be analyzed belongs from the candidate categories. This application optimizes ultrasonic data before the measurement statistics of the ultrasonic flow meter. It allows for multi-dimensional analysis of the feature content of the ultrasonic data to be analyzed, even without any ultrasonic data of known category. This allows for precise determination of the ultrasonic data category, thus enabling more accurate optimization of the ultrasonic data and improving the measurement efficiency of the ultrasonic flow meter. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating a data optimization method for an ultrasonic flow meter provided in an embodiment of this application. Detailed Implementation

[0022] To better understand the above technical solutions, the technical solutions of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of this application and the specific features in the embodiments are detailed descriptions of the technical solutions of this application, rather than limitations on the technical solutions of this application. In the absence of conflict, the embodiments of this application and the technical features in the embodiments can be combined with each other.

[0023] Please see Figure 1 This paper illustrates a data optimization method for an ultrasonic flow meter, which may include the technical solutions described in steps S202-S210.

[0024] S202, extract the feature content of the ultrasonic data to be analyzed, and obtain the bias of each feature content.

[0025] The ultrasonic data to be analyzed is ultrasonic data whose type needs to be determined. The ultrasonic data to be analyzed can be ultrasonic data acquired by multiple ultrasonic devices. In this application, factors affecting the performance of the ultrasonic flowmeter include the geometry of the meter body, the position of the ultrasonic transducer, the uncertainty of known parameters (including temperature coefficient and pressure coefficient), and interference factors such as the accuracy and quality (e.g., the stability of the crystal oscillator) of the electronic components used in the transducer and transit time measurement circuit. Therefore, it is necessary to optimize the ultrasonic data to minimize interference.

[0026] Furthermore, the feature content can be a representative feature in the ultrasonic data to be analyzed, which can be used to represent the core content of the ultrasonic data to be analyzed (for example, the amplitude of the ultrasonic wave represents the energy transmitted in this ultrasonic wave, and when the ultrasonic waveform data is converted into an electrical signal, different energy levels represent a specific data or feature content).

[0027] The bias of a feature can be used to represent the importance of that feature in the ultrasound data to be analyzed. The bias can be understood as a weight, and in this application, it specifically represents the degree of importance of the feature in the ultrasound data.

[0028] S204, obtain the attribute characterization data corresponding to each feature content.

[0029] Attribute characterization data is data used to assist in understanding the description of features. Attribute characterization data can be in the form of electrical signals from ultrasonic data, with each signal corresponding to a specific feature.

[0030] In this embodiment, the attribute characterization data corresponding to the feature content can be determined based on information compiled by relevant technical personnel to describe the feature content (hereinafter referred to as expert description information). These technical personnel can be experts in the relevant field. Specifically, experts can compile their own corresponding expert description information for each candidate feature content. Then, through the relationships between each candidate feature content, each expert description information, and each candidate feature content and each expert description information, an expert knowledge base is built. Accordingly, when it is necessary to obtain the attribute characterization data of the feature content, the candidate feature content corresponding to that feature content is searched in the expert knowledge base. The attribute characterization data of the feature content can include the expert description information matching the found candidate feature content.

[0031] S206, by mapping the data of each attribute, determine the first common factor of each feature content with each candidate category.

[0032] The first common factor (which can be understood as similarity) between the feature content and the candidate category is an evaluation value used to assess the degree of association between the feature content and the candidate category. The value of the first common factor can range from 0% to 100%. The larger the first common factor, the higher the degree of association between the feature content and the candidate category, and vice versa.

[0033] In this embodiment, there are two or more candidate categories. The first common factor between each feature content and each candidate category is determined by the attribute characterization data corresponding to each feature content.

[0034] For example, the ultrasound data to be analyzed, a, is extracted into three features: feature a1, feature a2, and feature a3. Feature a1 corresponds to attribute characterization data a1a, feature a2 corresponds to attribute characterization data a2a, and feature a3 corresponds to attribute characterization data a3a. There are also three candidate categories: candidate category b1, candidate category b2, and candidate category b3.

[0035] Accordingly, based on attribute characterization data a1a, the first common factor between feature content a1 and candidate category b1, the first common factor between feature content a1 and candidate category b2, and the first common factor between feature content a1 and candidate category b3 are determined. Furthermore, based on attribute characterization data a2a, the first common factor between feature content a2 and candidate category b1, the first common factor between feature content a2 and candidate category b2, and the first common factor between feature content a2 and candidate category b3 are determined. Finally, based on attribute characterization data a3a, the first common factor between feature content a3 and candidate category b1, the first common factor between feature content a3 and candidate category b2, and the first common factor between feature content a3 and candidate category b3 are determined.

[0036] In one embodiment, the first common factor between each feature content and each candidate category can be determined using the attribute characterization data corresponding to each feature content and the category description data of each candidate category. For example, the first common factor between feature content a1 and candidate category b1 can be determined based on attribute characterization data a1a and category description data of candidate category b1. Here, the category description data of the candidate category is information that can be used to characterize the properties of that candidate category.

[0037] S208, by using the bias of each feature content and each first common factor, determine the second common factors of the ultrasonic data to be analyzed and each candidate type.

[0038] The second commonality factor between the ultrasound data to be analyzed and the candidate categories is an evaluation value used to assess the degree of correlation between the ultrasound data to be analyzed and the candidate categories. The value of the second commonality factor can range from 0% to 100%. The larger the second commonality factor, the higher the degree of correlation between the ultrasound data to be analyzed and the candidate categories; conversely, the smaller the second commonality factor, the lower the degree of correlation between the ultrasound data to be analyzed and the candidate categories.

[0039] In this embodiment, for each candidate category, a second common factor between the ultrasonic data to be analyzed and the candidate category is obtained by weighted summation based on the bias of each feature content of the ultrasonic data to be analyzed and the first common factor of each feature content with the candidate category.

[0040] Furthermore, a weighted sum is performed based on the bias of feature content a1, the first common factor between feature content a1 and candidate category b1, the bias of feature content a2, the first common factor between feature content a2 and candidate category b1, the bias of feature content a3, and the first common factor between feature content a3 and candidate category b1 to obtain the second common factor between the ultrasonic data a to be analyzed and candidate category b1. Additionally, a weighted sum is performed based on the bias of feature content a1, the first common factor between feature content a1 and candidate category b2, the bias of feature content a2, the first common factor between feature content a2 and candidate category b2, the bias of feature content a3, and the first common factor between feature content a3 and candidate category b2 to obtain the second common factor between the ultrasonic data a to be analyzed and candidate category b2. Furthermore, a weighted sum is performed based on the bias of feature content a1, the first common factor between feature content a1 and candidate category b3, the bias of feature content a2, the first common factor between feature content a2 and candidate category b3, the bias of feature content a3, and the first common factor between feature content a3 and candidate category b3 to obtain the second common factor between the ultrasonic data a to be analyzed and candidate category b3.

[0041] S210, using each of the second common factors, determine the type of the ultrasonic data to be analyzed from each of the candidate types, and perform optimization processing on the type of the ultrasonic data to be analyzed to determine the optimization result of the ultrasonic data type.

[0042] The category to which the ultrasound data to be analyzed belongs can include candidate categories that meet the common factor selection requirements of the second common factor of the ultrasound data to be analyzed. The common factor selection requirements can be set according to actual needs.

[0043] Specifically, the selection requirements for common factors may include: the second common factor with the ultrasound data to be analyzed is not less than a specified value of the common factor, which is predetermined according to actual needs. Alternatively, the selection requirements may include: the second common factor with the ultrasound data to be analyzed is among the pre-set number of second common factors with the largest value, i.e., sorted according to a certain order of the values ​​of the second common factors (which can be sorted from largest to smallest or smallest to largest). The category to which the ultrasound data to be analyzed belongs may include the candidate categories corresponding to the pre-set number of second common factors listed first, which can be set to any positive integer according to actual needs.

[0044] Understandably, a classification system (a system formed by defining and dividing the target domain) can be pre-defined, encompassing more than one candidate category. Based on this, the category to which the ultrasonic data to be analyzed belongs can be determined from the candidate categories covered by the classification system using each of the second commonality factors.

[0045] In one embodiment, after step S210, the following step may be further included: labeling the ultrasonic data to be analyzed according to its category. Specifically, the category labeling may involve outputting the category labeling result corresponding to the ultrasonic data to be analyzed. This category labeling result may include the ultrasonic data to be analyzed, the category to which the ultrasonic data to be analyzed belongs, and the common factors between the ultrasonic data to be analyzed and its category. The common factors between the ultrasonic data to be analyzed and its category can be determined based on a second common factor between the ultrasonic data to be analyzed and its category, such as the second common factor between the ultrasonic data to be analyzed and its category itself.

[0046] This application provides a data optimization method for ultrasonic flow meters. The method extracts feature content from the ultrasonic data to be analyzed and obtains the bias degree of each feature content. Then, it obtains attribute characterization data corresponding to each feature content. Using the attribute characterization data, it determines the first common factor between each feature content and the candidate categories. Then, using the bias degree of each feature content and the first common factor, it determines the second common factor between the ultrasonic data to be analyzed and each candidate category. Finally, using the second common factor, it determines the category to which the ultrasonic data to be analyzed belongs from the candidate categories. This application optimizes ultrasonic data before the metering statistics of the ultrasonic flow meter. Thus, even without any ultrasonic data whose category is known, it can perform multi-dimensional analysis of the feature content of the ultrasonic data to be analyzed, accurately determining the category of the ultrasonic data. This allows for more precise optimization of the ultrasonic data, thereby improving the metering efficiency of the ultrasonic flow meter.

[0047] In one embodiment, the step of extracting the feature content of the ultrasonic data to be analyzed may include the following steps: performing local feature processing on the ultrasonic data to be analyzed to obtain X first local features of the ultrasonic data to be analyzed; deleting first local features belonging to the target filtering feature set from each first local feature to determine Y second local features; obtaining the feature content of the ultrasonic data to be analyzed through each second local feature; the second local features include the first local features remaining after deletion; wherein, X is greater than 0 and X is greater than or equal to Y.

[0048] The filtering features recorded in the filtering feature set are the feature phrases that need to be deleted from each of the first local features of the ultrasound data to be analyzed. These features may include at least one of feature phrases that have no practical meaning and feature phrases commonly found in the ultrasound data covered by the corresponding initial data source. It is understood that feature phrases commonly found in the ultrasound data covered by the initial data source are difficult to represent the characteristics of a single ultrasound data point within that initial data source. Therefore, it is difficult to distinguish between the various ultrasound data points covered by that initial data source based on the feature phrase. Thus, the feature phrase can be used as a filtering feature of that initial data source and included in the filtering feature set corresponding to that initial data source.

[0049] The target filtering feature set is the set of filtering features corresponding to the target initial data source. The target initial data source is the initial data source to which the ultrasonic data to be analyzed belongs. In practice, the initial data source to which the ultrasonic data to be analyzed belongs (i.e., the target initial data source) can be determined first, and then the filtering feature set corresponding to that target initial data source (i.e., the target filtering feature set) can be determined.

[0050] In this embodiment, local feature processing can be performed on the ultrasonic data to be analyzed to obtain each first local feature of the ultrasonic data to be analyzed. Then, from each first local feature, the first local features that are the same as the filtering features recorded in the target filtering feature set are deleted. Then, based on the remaining first local features after deletion (i.e., second local features), the feature content of the ultrasonic data to be analyzed is obtained.

[0051] In one embodiment, the method of constructing a target filtering feature set may include the following steps: performing local feature processing on ultrasonic data belonging to the initial data source of the target to obtain several third local features; sequentially determining the first proportion corresponding to each third local feature; and constructing a target filtering feature set based on the third local features whose first proportion is greater than a specified value of the first proportion.

[0052] The first proportion corresponding to the third local feature can be the proportion of the ultrasonic dataset that covers the third local feature in the target initial data source to the total number of ultrasonic data in the target initial data source.

[0053] The first percentage is specified as a standard for evaluating whether the third local feature is a common feature among the ultrasound data covered by the target initial data source. If the first percentage corresponding to the third local feature is greater than the first percentage specified value, it indicates that the third local feature is a common feature among the ultrasound data covered by the target initial data source and should be used as a filtering feature of the target initial data source; if the first percentage corresponding to the third local feature is not greater than the first percentage specified value, it indicates that the third local feature is not a common feature among the ultrasound data covered by the target initial data source and should not be used as a filtering feature of the target initial data source.

[0054] In this embodiment, the target initial data source covers several ultrasonic data sets. Local feature processing can be performed on each ultrasonic data set covered by the target initial data source to obtain several third local features. For each third local feature, the proportion of the ultrasonic dataset covering that third local feature in the target initial data source relative to the total number of ultrasonic data sets in the target initial data source is determined to obtain a first proportion corresponding to each third local feature. Then, third local features with a first proportion greater than a specified first proportion are selected from the third local features. Based on the selected third local features, a target filtering feature set is constructed. Accordingly, the constructed target filtering feature set records the third local features with a first proportion greater than the specified first proportion.

[0055] In one embodiment, a target filtering feature set can also be constructed based on manually determined features that have no practical meaning for the target's initial data source, and third local features whose first proportion is greater than the specified first proportion value.

[0056] Understandably, constructing the filter feature sets corresponding to the initial data sources can be a pre-prepared task. Specifically, filter feature sets corresponding to each initial data source can be pre-constructed. After obtaining the paper to be annotated, the initial data source to which the ultrasonic data to be analyzed belongs (i.e., the target initial data source) is determined. When the filter feature set corresponding to the target initial data source (i.e., the target filter feature set) is needed, it can be directly found from the pre-constructed filter feature sets corresponding to each initial data source, without the need to construct the target filter feature set on a temporary basis. Furthermore, the filter feature sets corresponding to each initial data source can be periodically adjusted.

[0057] In one embodiment, in addition to determining the first proportion specification value through the artificial setting method described above, the first proportion specification value can also be determined using the following steps: S602, determining the fourth local feature from each third local feature based on the real-time proportion specification value; S604, determining the number of remaining features corresponding to each ultrasonic data belonging to the target initial data source; S606, determining the second proportion of the ultrasonic dataset with a number of remaining features not less than the specified feature number to the total number of ultrasonic data in the target initial data source; S608, when the second proportion is not greater than the second proportion specification value, determining the real-time proportion specification value as the first proportion specification value; S610, when the second proportion is greater than the second proportion specification value, adjusting the real-time proportion specification value according to the adjustment value, and returning to the step of determining the fourth local feature from each third local feature based on the real-time proportion specification value.

[0058] The fourth local feature may include a third local feature whose first proportion is not less than a specified real-time proportion. Specifically, the fourth local feature is the third local feature selected from the third local features obtained by performing local feature processing on each ultrasonic data belonging to the initial data source of the target, whose first proportion is not less than a specified real-time proportion.

[0059] The number of remaining features corresponding to the ultrasonic data can be the number of third local features remaining after deleting the fourth local feature from each of the third local features of the ultrasonic data.

[0060] The second specified percentage is used to evaluate whether the real-time specified percentage can be used as the first specified percentage. After determining the second percentage, it is checked whether the second percentage is greater than the second specified percentage. If it is not greater, the real-time specified percentage is determined as the first specified percentage, and the process of determining the first specified percentage ends. If it is greater, it indicates that the real-time specified percentage cannot be used as the first specified percentage. Then, the real-time specified percentage is adjusted according to the adjustment value. That is, the adjustment value is subtracted from the real-time specified percentage, and the step of determining the fourth local feature from each third local feature and subsequent steps are re-executed based on the adjusted real-time specified percentage. The second specified percentage can be determined according to actual needs, for example, it can be set to 80%.

[0061] Furthermore, an initial data source can correspond to a first specified percentage value. During the process of determining the first specified percentage value for an initial data source, when the real-time specified percentage value is determined for the first time, the initial specified percentage value is set as the real-time specified percentage value. The initial specified percentage value can be predetermined according to actual needs, for example, it can be set to 99%.

[0062] Understandably, compared to manually setting a first proportion specification value, this method determines a second proportion of ultrasonic data with a number of remaining features not less than a specified feature number, relative to the ultrasonic data belonging to the initial data source of the target. If the second proportion is determined to be greater than the second proportion specification value, the real-time proportion specification value is decreased, and the second proportion is determined again until it is no longer greater than the second proportion specification value. At this point, the real-time proportion specification value is set as the first proportion specification value. This achieves automatic determination of the first proportion specification value and improves the accuracy of the determined first proportion specification value.

[0063] In one embodiment, the step of obtaining the feature content of the ultrasonic data to be analyzed through each second local feature may include the following steps: distributing the second local features to obtain a fifth local feature; determining a sixth local feature from each fifth local feature; determining a seventh local feature from each sixth local feature; and obtaining the feature content of the ultrasonic data to be analyzed based on the seventh local feature.

[0064] The fifth local feature encompasses at least two consecutive second local features. Specifically, after deleting the first local features belonging to the target filtering feature set from each first local feature, the remaining first local features (i.e., each second local feature) are distributed according to a pre-defined distribution rule to obtain all combined features that encompass at least two consecutive second local features. Each combined feature is the fifth local feature.

[0065] The sixth local feature can be a key feature that belongs to the fifth local feature.

[0066] The seventh local feature is not covered by the content corresponding to each of the sixth local features.

[0067] In one embodiment, the feature content of the ultrasonic data to be analyzed is obtained based on the seventh local feature. Specifically, the seventh local feature may be determined as the feature content of the ultrasonic data to be analyzed.

[0068] In one embodiment, the step of obtaining the attribute characterization data corresponding to each feature content, i.e., step S204, may include the following steps: obtaining the real-time ultrasonic depiction content corresponding to each feature content; and obtaining the attribute characterization data corresponding to each feature content through the real-time ultrasonic depiction content corresponding to each feature content.

[0069] In one embodiment, attribute characterization data corresponding to the feature content is obtained based on the real-time ultrasonic characterization content corresponding to the feature content. Specifically, the attribute characterization data corresponding to the feature content may include the real-time ultrasonic characterization content corresponding to the feature content.

[0070] In one embodiment, the step of obtaining real-time ultrasound mapping content corresponding to each feature content may include the following steps: searching for candidate feature content corresponding to each feature content in the database; the database records the association between candidate feature content and candidate matching information, and the candidate matching information is obtained by matching the corresponding candidate feature content through a real-time data matching unit; when a candidate feature content corresponding to a feature content is found, obtaining the real-time ultrasound mapping content corresponding to that feature content based on the candidate matching information matched by the found candidate feature content; when no candidate feature content corresponding to a feature content is found, calling the real-time data matching unit to match the feature content to obtain the real-time ultrasound mapping content corresponding to that feature content.

[0071] In this embodiment, the real-time data matching unit can be called in advance to match each candidate feature content. The target matching results corresponding to each candidate feature content are the candidate matching information. Then, a database is generated that records each candidate feature content, each candidate matching information, and the association between each candidate feature content and each candidate matching information. The contents of the database are then stored.

[0072] When it is necessary to obtain real-time ultrasound mapping content corresponding to the feature content, the candidate feature content corresponding to the feature content can be directly searched in the database. If the candidate feature content corresponding to the feature content is found in the database, it means that the feature content has been matched by the real-time data matching unit in advance. At this time, it is not necessary to call the real-time data matching unit to match the feature content again. Instead, the candidate matching information matched by the candidate feature content corresponding to the feature content can be directly used as the real-time ultrasound mapping content corresponding to the feature content.

[0073] Conversely, if no candidate feature content corresponding to the given feature content is found in the database, it indicates that the real-time data matching unit has not been invoked to match the feature content beforehand, and therefore the database does not store candidate matching information that can serve as the real-time ultrasound mapping content corresponding to the feature content. In this case, the real-time data matching unit can be temporarily invoked to match the feature content, and the target matching result corresponding to the feature content is the real-time ultrasound mapping content corresponding to the feature content. Furthermore, the feature content and the target matching result corresponding to the feature content can be used as newly added candidate feature content and newly added candidate matching information, and debugged into a database that records each candidate feature content, each candidate matching information, and the association between each candidate feature content and each candidate matching information.

[0074] Understandably, the real-time data matching unit is invoked only when no matching candidate feature is found in the database. This significantly improves the efficiency of determining the type of ultrasound data to be analyzed. Furthermore, in practical applications, the number of frequently occurring features in ultrasound data is relatively limited. Once the database accumulates tens of millions of candidate features, it becomes much less necessary to call an external real-time data matching unit to obtain the corresponding real-time ultrasound description. This allows for highly efficient determination of the type of massive amounts of ultrasound data.

[0075] In addition, the database, which records the content of each candidate feature, each candidate matching information, and the relationship between each candidate feature and each candidate matching information, can be debugged periodically.

[0076] In one embodiment, the step of obtaining attribute characterization data corresponding to each feature content through the real-time ultrasonic depiction content corresponding to each feature content may include the following steps: filtering the real-time ultrasonic depiction content corresponding to each feature content to obtain attribute characterization data corresponding to each feature content.

[0077] Data filtering can remove information that is irrelevant to the feature content itself from the real-time ultrasonic mapping content corresponding to the feature content.

[0078] In one embodiment, the step of determining the first common factor between each feature content and each candidate category through each attribute characterization data, i.e., step S206, may include the following steps: determining the third common factor between each attribute characterization data and each candidate category through each attribute characterization data and the category label of each candidate category; and determining the first common factor between each feature content and each candidate category through each third common factor.

[0079] The third common factor between attribute characterization data and candidate species is determined based on the attribute characterization data and the species label of the candidate species. It is an evaluation value used to assess the degree of association between the attribute characterization data and the candidate species. The value of the third common factor can range from 0% to 100%. The larger the third common factor, the higher the degree of association between the attribute characterization data and the candidate species based on the species label of the candidate species. Conversely, the smaller the third common factor, the lower the degree of association between the feature content and the candidate species based on the attribute characterization data and the species label of the candidate species.

[0080] For each feature, a first common factor between the feature and each candidate species can be determined based on the attribute characterization data corresponding to the feature and the species description data of each candidate species. In this embodiment, the species description data of the candidate species may include the species label of the candidate species. Accordingly, for each feature, a third common factor between the attribute characterization data corresponding to the feature and each candidate species can be determined based on the attribute characterization data corresponding to the feature and the species label of each candidate species. Furthermore, a first common factor between the feature and each candidate species can be determined based on the third common factor between the attribute characterization data corresponding to the feature and each candidate species.

[0081] The feature contents of the ultrasonic data a to be analyzed are feature content a1, feature content a2 and feature content a3. Feature content a1 corresponds to attribute characterization data a1a, feature content a2 corresponds to attribute characterization data a2a and feature content a3 corresponds to attribute characterization data a3a. The candidate categories are candidate category b1, candidate category b2 and candidate category b3.

[0082] Based on this, the third common factor between attribute characterization data a1a and candidate category b1 is determined according to the category label of the attribute characterization data a1a and candidate category b1. Then, based on the third common factor between attribute characterization data a1a and candidate category b1, the first common factor between feature content a1 and candidate category b1 is determined.

[0083] Based on the attribute characterization data a1a and the category label of the candidate category b2, the third common factor between the attribute characterization data a1a and the candidate category b2 is determined. Then, based on the third common factor between the attribute characterization data a1a and the candidate category b2, the first common factor between the feature content a1 and the candidate category b2 is determined.

[0084] Based on the attribute characterization data a1a and the category label of the candidate category b3, the third common factor between the attribute characterization data a1a and the candidate category b3 is determined. Then, based on the third common factor between the attribute characterization data a1a and the candidate category b3, the first common factor between the feature content a1 and the candidate category b3 is determined.

[0085] By analogy, the first common factor of feature content a2 with candidate categories b1, b2 and b3 is determined, and the first common factor of feature content a3 with candidate categories b1, b2 and b3 is determined.

[0086] In one embodiment, the third common factor between the attribute characterization data corresponding to the feature content and the candidate category is the first common factor between the feature content and the candidate category. For example, the third common factor between attribute characterization data a1a and candidate category b1 is the first common factor between feature content a1 and candidate category b1.

[0087] In one embodiment, the step of determining the third common factor between each attribute characterization data and each candidate species using attribute characterization data and category labels of each candidate species may include the following steps: determining the intersection features between each attribute characterization data and each candidate species' category labels using attribute characterization data and each candidate species' category labels; determining the target intersection features between each attribute characterization data and each candidate species' category labels from the intersection features between attribute characterization data and each candidate species' category labels; determining the third ratio of the total number of features of the target intersection features between attribute characterization data and each candidate species' category labels to the total number of features of the category labels of each candidate species; and determining the third common factor between each attribute characterization data and each candidate species' category labels using the third ratio, the first feature point of the target intersection features between attribute characterization data and each candidate species' category labels in the corresponding attribute characterization data, and the first abnormal text probability of the target intersection features between attribute characterization data and each candidate species' category labels.

[0088] In this embodiment, for each attribute characterization data, the intersection features between the attribute characterization data and the category label of each candidate category are determined sequentially. For example, if there are three attribute characterization data: attribute characterization data a1a, attribute characterization data a2a, and attribute characterization data a3a, and three candidate categories: candidate category b1, candidate category b2, and candidate category b3, then the intersection features between attribute characterization data a1a and candidate category b1, the intersection features between attribute characterization data a1a and candidate category b2, and the intersection features between attribute characterization data a1a and candidate category b3 are determined. Similarly, the intersection features between attribute characterization data a2a and candidate categories b1, b2, and b3 are determined, as are the intersection features between attribute characterization data a3a and candidate categories b1, b2, and b3.

[0089] The target intersection features of the attribute characterization data and the category label of the candidate species are not included in the intersection features of the attribute characterization data and the category label of the candidate species other than the intersection features of the attribute characterization data and the category label of the candidate species.

[0090] In this embodiment, for each attribute characterization data, the target intersection features between the attribute characterization data and the category label of each candidate category are determined from the intersection features between the attribute characterization data and the category label of each candidate category.

[0091] Furthermore, from the intersection features of attribute characterization data a1a and candidate type b1, the target intersection features of attribute characterization data a1a and candidate type b1 are determined; from the intersection features of attribute characterization data a1a and candidate type b2, the target intersection features of attribute characterization data a1a and candidate type b2 are determined; and from the intersection features of attribute characterization data a1a and candidate type b3, the target intersection features of attribute characterization data a1a and candidate type b3 are determined.

[0092] From the intersection features of attribute characterization data a2a with candidate categories b1, b2 and b3 respectively, the target intersection features of attribute characterization data a2a with candidate categories b1, b2 and b3 are determined.

[0093] From the intersection features of attribute characterization data a3a with candidate categories b1, b2 and b3 respectively, the target intersection features of attribute characterization data a3a with candidate categories b1, b2 and b3 are determined.

[0094] The third proportion is the percentage of the total number of features of the target intersection features between the attribute characterization data and the category labels of the candidate category to the total number of features of the category labels of the candidate category.

[0095] In this embodiment, for each attribute characterization data, the third proportion of the total number of features of the target intersection features between the attribute characterization data and the category label of each candidate category is determined.

[0096] The total number of features of the target intersection features between attribute characterization data a1a and candidate species b1 is determined as the third percentage of the total number of features of the category label of candidate species b1; the total number of features of the target intersection features between attribute characterization data a1a and candidate species b2 is determined as the third percentage of the total number of features of the category label of candidate species b2; the total number of features of the target intersection features between attribute characterization data a1a and candidate species b3 is determined as the third percentage of the total number of features of the category label of candidate species b3.

[0097] The total number of features of the intersection features between attribute characterization data a2a and the target features of candidate categories b1, b2 and b3 is determined to be the third proportion of the total number of features of the category labels of candidate categories b1, b2 and b3.

[0098] The total number of features of the intersection features between attribute characterization data a3a and the target features of candidate categories b1, b2 and b3 is determined to be the third proportion of the total number of features of the category labels of candidate categories b1, b2 and b3.

[0099] The first feature point of the target intersection feature between the attribute characterization data and the category label of the candidate category in the attribute characterization data is the number of times the target intersection feature appears in the attribute characterization data.

[0100] In this embodiment, for each attribute characterization data, the third commonality factor between the attribute characterization data and the category label of each candidate category is determined based on the third proportion of the total number of features of the target intersection features of the attribute characterization data and the category label of each candidate category to the total number of features of the category label of each candidate category, the first feature point of the target intersection features of the attribute characterization data and the category label of each candidate category in the attribute characterization data, and the first abnormal text probability of the target intersection features of the attribute characterization data and the category label of each candidate category.

[0101] In one embodiment, a data optimization method for an ultrasonic flow meter may further include the following steps: determining a fourth common factor between each attribute characterization data and each candidate type by using the attribute characterization data, and the pre-set type association features and corresponding pre-set related parameters of each candidate type. Accordingly, the step of determining a first common factor between each feature content and each candidate type by using the third common factor may include the following steps: determining a first common factor between each feature content and each candidate type by using the third common factor and the fourth common factor.

[0102] The pre-defined category association features of the candidate categories are manually determined features that are related to the candidate categories. The pre-defined category association features and the relevant parameters of the candidate categories are used to represent the correlation between the pre-defined category association features and the candidate categories. Specifically, the pre-defined category association features of the candidate categories, and the relevant parameters of the pre-defined category association features and the candidate categories, can be pre-determined manually based on experience accumulated in actual business operations.

[0103] The fourth common factor between attribute characterization data and candidate species is determined based on the attribute characterization data, the pre-set species association features of the candidate species, and the relevant parameters of the candidate species. It is an evaluation value used to assess the degree of association between the attribute characterization data and the candidate species. The value of the fourth common factor can range from 0% to 100%. The larger the fourth common factor, the higher the degree of association between the attribute characterization data and the candidate species based on the pre-set species association features of the candidate species and the relevant parameters of the candidate species. Conversely, the smaller the fourth common factor, the lower the degree of association between the attribute characterization data and the candidate species based on the pre-set species association features of the candidate species and the relevant parameters of the candidate species.

[0104] Furthermore, for each feature content, a first common factor between the feature content and each candidate species can be determined based on the attribute characterization data corresponding to the feature content and the species description data of each candidate species. In this embodiment, the species description data of the candidate species may include pre-set species association features of the candidate species and pre-set related parameters of the pre-set species association features and the corresponding candidate species. For each feature content, a fourth common factor between the attribute characterization data corresponding to the feature content and each candidate species is determined based on the attribute characterization data corresponding to the feature content, the pre-set species association features of each candidate species, and the related parameters of the pre-set species association features of each candidate species and their respective corresponding candidate species. Then, a first common factor between the feature content and each candidate species is determined based on the third common factor between the attribute characterization data corresponding to the feature content and each candidate species, and the fourth common factor between the attribute characterization data corresponding to the feature content and each candidate species.

[0105] For example, the feature contents of the ultrasound data a to be analyzed are feature content a1, feature content a2 and feature content a3. Feature content a1 corresponds to attribute characterization data a1a, feature content a2 corresponds to attribute characterization data a2a, and feature content a3 corresponds to attribute characterization data a3a. The candidate categories are candidate category b1, candidate category b2 and candidate category b3.

[0106] Accordingly, based on the attribute characterization data a1a and the category label of the candidate category b1, the third common factor between the attribute characterization data a1a and the candidate category b1 is determined. Based on the pre-set category association features of the attribute characterization data a1a and the candidate category b1, and the relevant parameters of the pre-set category association features and the candidate category b1, the fourth common factor between the attribute characterization data a1a and the candidate category b1 is determined. Then, based on the third common factor between the attribute characterization data a1a and the candidate category b1, and the fourth common factor between the attribute characterization data a1a and the candidate category b1, the first common factor between the feature content a1 and the candidate category b1 is jointly determined.

[0107] Based on the attribute characterization data a1a and the category label of the candidate category b2, the third common factor between attribute characterization data a1a and candidate category b2 is determined. Based on the pre-set category association features of attribute characterization data a1a and candidate category b2, and the relevant parameters of the pre-set category association features and candidate category b2, the fourth common factor between attribute characterization data a1a and candidate category b2 is determined. Then, based on the third common factor between attribute characterization data a1a and candidate category b2 and the fourth common factor between attribute characterization data a1a and candidate category b2, the first common factor between feature content a1 and candidate category b2 is determined.

[0108] Based on the attribute characterization data a1a and the category label of the candidate category b3, the third common factor between attribute characterization data a1a and candidate category b3 is determined. Based on the pre-set category association features of attribute characterization data a1a and candidate category b3, and the relevant parameters of the pre-set category association features and candidate category b3, the fourth common factor between attribute characterization data a1a and candidate category b3 is determined. Then, based on the third common factor between attribute characterization data a1a and candidate category b3 and the fourth common factor between attribute characterization data a1a and candidate category b3, the first common factor between feature content a1 and candidate category b3 is determined.

[0109] By analogy, the first common factor of feature content a2 with candidate categories b1, b2, and b3 is determined. Furthermore, the first common factor of feature content a3 with candidate categories b1, b2, and b3 is determined.

[0110] Specifically, for any feature content and any candidate category, the first common factor between the feature content and the candidate category can be obtained by performing a simple summation on the attribute characterization data corresponding to the feature content and the third common factor of the candidate category, as well as the fourth common factor of the attribute characterization data corresponding to the feature content and the candidate category. For example, the first common factor between feature content a1 and candidate category b1 can be obtained by performing a simple summation on the third common factor of attribute characterization data a1a and candidate category b1, and the fourth common factor of attribute characterization data a1a and candidate category b1.

[0111] Alternatively, biases can be set for the third and fourth common factors respectively. A weighted sum can be calculated based on the attribute characterization data corresponding to the feature content, the third common factor of the candidate category, the bias of the third common factor, the attribute characterization data corresponding to the feature content, the fourth common factor of the candidate category, and the bias of the fourth common factor, to obtain the first common factor between the feature content and the candidate category. For example, the first common factor between feature content a1 and candidate category b1 can be obtained by weighted summing the attribute characterization data a1a, the third common factor of the candidate category b1, the bias of the third common factor, the fourth common factor of the candidate category b1, and the bias of the fourth common factor.

[0112] In practice, a knowledge base can be pre-built, containing several manually determined related knowledge entries. Based on this knowledge base, pre-defined category association features and corresponding pre-defined related parameters for each candidate category can be obtained. Then, using the attribute characterization data, the pre-defined category association features, and corresponding pre-defined related parameters for each candidate category, the fourth common factor between each attribute characterization data and each candidate category can be determined.

[0113] In other embodiments, the first common factor of each feature content with each candidate category can be determined by comparing each attribute characterization data with the fourth common factor of each candidate category, without considering the third common factor of each attribute characterization data with each candidate category.

[0114] In one embodiment, the step of determining the fourth common factor between each attribute characterization data and each candidate type using attribute characterization data, pre-set type association features of each candidate type, and pre-set correlation parameters of the pre-set type association features of each candidate type and the corresponding candidate type may include the following steps: determining the fourth common factor between each attribute characterization data and each candidate type using the second feature points of the pre-set type association features of each candidate type in each attribute characterization data, the second abnormal text probability of the pre-set type association features of each candidate type, and the pre-set correlation parameters of the pre-set type association features of each candidate type and the corresponding candidate type.

[0115] In this embodiment, for each attribute characterization data, the fourth commonality factor between the attribute characterization data and the category label of each candidate category is determined based on the second feature point of each candidate category's pre-set category association feature in the attribute characterization data, the second abnormal text probability of each candidate category's pre-set category association feature, and the relevant parameters of each candidate category's pre-set category association feature with its corresponding candidate category.

[0116] The feature points of the feature content in the ultrasonic data to be analyzed represent the frequency of occurrence of the feature content in the ultrasonic data. The probability of abnormal text in the feature content can be understood as the likelihood of abnormal data. Specifically, the first feature point can be understood as ultrasonic wavelength information, and the second feature point can be understood as ultrasonic amplitude information; the first probability of abnormal text can be understood as the probability of abnormal generation of ultrasonic wavelength information, and the second probability of abnormal text can be understood as the probability of abnormal generation of ultrasonic amplitude information.

[0117] Understandably, after calling the real-time data matching unit to match the feature content, the number of all matching results obtained for that feature content and the attribute characterization data corresponding to that feature content can be obtained simultaneously, and the two can be correlated. Therefore, when obtaining the attribute characterization data corresponding to the feature content, the first parameter—the number of all matching results obtained for that feature content—can be obtained simultaneously. When calculating the abnormal text probability of the feature content, the first parameter can be directly used without temporarily calling the real-time data matching unit to obtain it.

[0118] Based on the network description information corresponding to each feature content of the ultrasonic data to be analyzed, the ultrasonic data description information corresponding to each feature content is determined. Furthermore, by comparing each attribute characterization data with at least one of the third common factor of each candidate category (determined by each attribute characterization data and the category label of each candidate category) and each attribute characterization data with each candidate category's fourth common factor (determined by each attribute characterization data, pre-set category association features of each candidate category, and pre-set related parameters of the corresponding candidate category), the second common factor between the ultrasonic data to be analyzed and each candidate category is determined. Then, analysis is performed based on the second common factor between the ultrasonic data to be analyzed and each candidate category. A basic assumption exists: by matching the feature content extracted from the ultrasonic data through a real-time data matching unit, if the label or category association feature of a certain candidate category frequently appears in the obtained matching results, then the ultrasonic data is closely related to that candidate category.

[0119] In one embodiment, a data optimization method for an ultrasonic flow meter may further include the following steps: obtaining importance level parameters for each candidate category. Accordingly, the step of determining the category to which the ultrasonic data to be analyzed belongs from each candidate category using each second common factor, i.e., step S210, may include the following steps: determining the fifth common factor between the ultrasonic data to be analyzed and each candidate category using each second common factor and the importance level parameters for each candidate category; determining the category to which the ultrasonic data to be analyzed belongs from each candidate category using each fifth common factor.

[0120] In this embodiment, for each candidate category, a fifth common factor is determined based on the second common factor between the ultrasonic data to be analyzed and the candidate category, and the importance level parameter of the candidate category. Specifically, the fifth common factor of the candidate category can be obtained by multiplying the second common factor between the ultrasonic data to be analyzed and the candidate category by the importance level parameter of the candidate category.

[0121] Furthermore, when outputting the category labeling results corresponding to the ultrasonic data to be analyzed, the common factor between the ultrasonic data to be analyzed and its category in the category labeling results can be the fifth common factor between the ultrasonic data to be analyzed and its category itself.

[0122] One embodiment provides a data optimization method for an ultrasonic flow meter. Specifically, this method may include the following steps S1202 to S1224.

[0123] S1202, extract the feature content of the ultrasonic data to be analyzed, and determine the bias of each feature content.

[0124] S1204, search for the candidate feature content corresponding to each feature content in the database; the database records the association between the candidate feature content and the candidate matching information, and the candidate matching information is obtained by the real-time data matching unit matching the corresponding candidate feature content.

[0125] S1206, when a candidate feature content corresponding to the feature content is found, the real-time ultrasonic depiction content corresponding to the feature content is obtained according to the candidate matching information matched by the found candidate feature content.

[0126] S1208, when no candidate feature content corresponding to the feature content is found, the real-time data matching unit is called to match the feature content to obtain the real-time ultrasonic depiction content corresponding to the feature content.

[0127] S1210, respectively filter the real-time ultrasonic depiction content corresponding to each feature content to obtain the attribute depiction data corresponding to each feature content.

[0128] S1212, using the attribute characterization data and the category labels of each candidate category, determine the third common factor between each attribute characterization data and each candidate category.

[0129] S1214, through the attribute characterization data, and the pre-set type association characteristics of each candidate type and the pre-set related parameters of the corresponding candidate type, determine the fourth common factor of each attribute characterization data and each candidate type; wherein, the pre-set type association characteristics of the candidate type and the pre-set type association characteristics of the candidate type and the related parameters of the candidate type are determined manually.

[0130] S1216, through each third common factor and each fourth common factor, determine the first common factor of each feature content and each candidate category respectively.

[0131] S1218, by using the bias of each feature content and each first common factor, determine the second common factors of the ultrasonic data to be analyzed and each candidate type.

[0132] S1220, obtain the important level parameters of each candidate category, and determine the fifth common factor of the ultrasound data to be analyzed and each candidate category through the second common factor and the important level parameters of each candidate category.

[0133] S1222, using each fifth common factor, determine the category to which the ultrasound data to be analyzed belongs from each candidate category.

[0134] S1224, Output the category labeling result corresponding to the ultrasonic data to be analyzed. The category labeling result corresponding to the ultrasonic data to be analyzed includes the ultrasonic data to be analyzed, the category to which the ultrasonic data to be analyzed belongs, and the fifth common factor between the ultrasonic data to be analyzed and the category to which the ultrasonic data to be analyzed belongs.

[0135] Based on the above, a data optimization device 200 for an ultrasonic flow meter is provided, the device comprising:

[0136] The feature determination module 210 is used to extract the feature content of the ultrasonic data to be analyzed and determine the bias degree of each feature content. The feature content is the feature in the ultrasonic data to be analyzed that represents the core content of the ultrasonic data to be analyzed. The bias degree indicates the importance of the corresponding feature content to the ultrasonic data to be analyzed.

[0137] The data characterization module 220 is used to obtain attribute characterization data corresponding to each of the aforementioned feature contents, wherein the attribute characterization data is used to assist in understanding the data described by the feature contents;

[0138] The first common factor determination module 230 is used to determine the first common factor between each of the feature contents and each of the candidate categories by using the attribute characterization data and the category description data of each candidate category. The category description data is used to characterize the characteristics of the candidate category. The first common factor is an evaluation value used to assess the degree of correlation between the feature contents and the candidate category.

[0139] The second common factor determination module 240 is used to determine the second common factor between the ultrasound data to be analyzed and each of the candidate categories by using the bias degree of each of the feature contents and each of the first common factors. The second common factor is an evaluation value used to assess the degree of correlation between the ultrasound data to be analyzed and the candidate categories.

[0140] The result optimization module 250 is used to determine the type of the ultrasonic data to be analyzed from the candidate types through each of the second common factors, and to optimize the type of the ultrasonic data to be analyzed to determine the optimization result of the ultrasonic data type.

[0141] Based on the above, a data optimization system 300 for an ultrasonic flow meter is shown, including a processor 310 and a memory 320 that communicate with each other. The processor 310 is used to read a computer program from the memory 320 and execute it to implement the above method.

[0142] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method during runtime.

[0143] In summary, based on the above scheme, the feature content of the ultrasonic data to be analyzed is extracted, and the bias degree of each feature content is obtained. Then, attribute characterization data corresponding to each feature content is obtained. Using the attribute characterization data, the first common factor between each feature content and the candidate categories is determined. Then, using the bias degree of each feature content and the first common factor, the second common factor between the ultrasonic data to be analyzed and each candidate category is determined. Finally, using the second common factor, the category to which the ultrasonic data to be analyzed belongs is determined from the candidate categories. This application optimizes the ultrasonic data before the measurement statistics of the ultrasonic flow meter. Thus, even without any ultrasonic data whose category is known, a multi-dimensional analysis of the feature content of the ultrasonic data to be analyzed can be performed, accurately determining the category of the ultrasonic data. This allows for more precise optimization of the ultrasonic data, thereby improving the measurement efficiency of the ultrasonic flow meter.

[0144] It should be understood that the systems and modules described above can be implemented in various ways. For example, in some embodiments, the systems and modules can be implemented by hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the methods and systems described above can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and modules of this application can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, or by a combination of the aforementioned hardware circuits and software (e.g., firmware).

[0145] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above, or any other possible beneficial effects.

[0146] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this application. Such modifications, improvements, and corrections are suggested in this application, and therefore remain within the spirit and scope of the exemplary embodiments of this application.

[0147] Furthermore, this application uses specific terms to describe embodiments of the application. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of the application. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of the application can be appropriately combined.

[0148] Furthermore, those skilled in the art will understand that aspects of this application can be described and illustrated through several patentable types or situations, including any new and useful combination of processes, machines, products, or substances, or any new and useful improvements thereof. Accordingly, aspects of this application can be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. All of the above hardware or software may be referred to as a “data block,” “module,” “engine,” “unit,” “component,” or “system.” Furthermore, aspects of this application may manifest as a computer product located on one or more computer-readable media, the product including computer-readable program code.

[0149] Computer storage media may contain a propagated data signal containing computer program code, for example, on baseband or as part of a carrier wave. This propagated signal may take various forms, including electromagnetic, optical, and suitable combinations thereof. Computer storage media can be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, apparatus, or device to enable communication, propagation, or transmission of a program for use. The program code located on the computer storage medium can be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0150] The computer program code required for the operation of each part of this application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby, and Groovy, or other programming languages. This program code can run entirely on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as Software as a Service (SaaS).

[0151] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this application are not intended to limit the order of the processes and methods of this application. Although the foregoing disclosure has discussed some currently considered useful embodiments of the invention through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments of this application. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely through software solutions, such as installing the described system on existing servers or mobile devices.

[0152] Similarly, it should be noted that, in order to simplify the description of the present application and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments of the present application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of the application requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0153] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are open to adaptive variation. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters are taken into account a specified number of significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of application in some embodiments of this application are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0154] For each patent, patent application, patent application publication, and other material such as articles, books, specifications, publications, and documents referenced in this application, the entire contents of that patent are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this application, as well as documents that limit the broadest scope of the claims in this application (currently or subsequently appended to this application). It should be noted that if there are any inconsistencies or conflicts between the descriptions, definitions, and / or terminology used in the supplementary materials of this application and the content of this application, the descriptions, definitions, and / or terminology used in this application shall prevail.

[0155] Finally, it should be understood that the embodiments described in this application are merely illustrative of the principles of the embodiments of this application. Other modifications may also fall within the scope of this application. Therefore, alternative configurations of the embodiments of this application are considered as examples and not limitations, and are regarded as consistent with the teachings of this application. Accordingly, the embodiments of this application are not limited to the embodiments explicitly described and illustrated in this application.

[0156] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A data optimization method for an ultrasonic flow meter, characterized in that, The method includes: The feature content of the ultrasound data to be analyzed is extracted, and the bias of each feature content is determined. The feature content is the feature in the ultrasound data to be analyzed that represents the core content of the ultrasound data to be analyzed. The bias indicates the importance of the corresponding feature content to the ultrasound data to be analyzed. Obtain attribute characterization data corresponding to each of the aforementioned feature contents, wherein the attribute characterization data is used to assist in understanding the data described by the feature contents; By using the attribute characterization data and the category description data of each candidate category, a first common factor is determined between each feature content and each candidate category. The category description data is used to characterize the characteristics of the candidate category. The first common factor is an evaluation value used to assess the degree of correlation between the feature content and the candidate category. By using the bias of each of the aforementioned features and each of the first common factors, the second common factors of the ultrasound data to be analyzed and each of the candidate categories are determined. The second common factor is an evaluation value used to assess the degree of correlation between the ultrasound data to be analyzed and the candidate categories. For each candidate category, the second common factor between the ultrasound data to be analyzed and the candidate category is obtained by weighted summation based on the bias of each feature content of the ultrasound data to be analyzed and the first common factor of each feature content. By using each of the second common factors, the type to which the ultrasonic data to be analyzed belongs is determined from each of the candidate types, and the type to which the ultrasonic data to be analyzed belongs is optimized to determine the optimization result of the ultrasonic data type. It also includes: obtaining the importance level parameters for each of the candidate categories; The step of determining the category to which the ultrasonic data to be analyzed belongs from each of the candidate categories using each of the second common factors, and optimizing the category to which the ultrasonic data to be analyzed belongs to determine the optimization result of the ultrasonic data category, includes: determining the fifth common factor between the ultrasonic data to be analyzed and each of the candidate categories using each of the second common factors and the importance level parameter of each of the candidate categories; determining the category to which the ultrasonic data to be analyzed belongs from each of the candidate categories using each of the fifth common factors, and optimizing the category to which the ultrasonic data to be analyzed belongs to determine the optimization result of the ultrasonic data category.

2. The method as described in claim 1, characterized in that, The extracted features of the ultrasonic data to be analyzed include: Local feature processing is performed on the ultrasonic data to be analyzed to obtain X first local features of the ultrasonic data to be analyzed. Deleting the first local features belonging to the target filtering feature set from each of the first local features, Y second local features are determined; the second local features include the remaining first local features after deletion; wherein, X is greater than 0, and X is greater than or equal to Y; The feature content of the ultrasonic data to be analyzed is obtained through each of the second local features; wherein, the target filter feature set includes a filter feature set corresponding to the target initial data source, and the target initial data source includes the initial data source to which the ultrasonic data to be analyzed belongs.

3. The method as described in claim 2, characterized in that, The methods for constructing the target filtering feature set include: Local feature processing is performed on each ultrasonic data belonging to the initial data source of the target to obtain several third local features; The first proportion corresponding to each of the third local features is determined sequentially; The first proportion corresponding to the third local feature includes: The percentage of the total number of ultrasound data in the target initial data source that includes the third local feature; The target filtering feature set is constructed by combining the third local feature whose first proportion is greater than the specified value of the first proportion.

4. The method as described in claim 3, characterized in that, The methods for determining the first specified percentage value include: Based on the specified real-time percentage value, a fourth local feature is determined from each of the third local features; the fourth local feature includes the third local feature whose first percentage is not less than the specified real-time percentage value. Determine the number of remaining features corresponding to each ultrasonic data belonging to the target initial data source; the number of remaining features corresponding to the ultrasonic data is the number of third local features remaining after deleting the fourth local feature from each of the third local features of the ultrasonic data; The ultrasonic dataset with the remaining feature count not less than a specified feature count is determined to be the second proportion of the total number of ultrasonic data belonging to the target initial data source; When the second percentage is not greater than the second percentage specified value, the real-time percentage specified value is determined as the first percentage specified value; When the second proportion is greater than the specified value of the second proportion, the specified value of the real-time proportion is adjusted according to the adjustment value, and the step of determining the fourth local feature from each of the third local features according to the specified value of the real-time proportion is returned.

5. The method as described in claim 2, characterized in that, The step of obtaining the feature content of the ultrasonic data to be analyzed through each of the second local features includes: The fifth local feature is obtained by distributing each of the second local features; Each of the fifth local features encompasses at least two consecutive second local features; from each of the fifth local features, a sixth local feature is determined; the sixth local feature includes key features belonging to the fifth local features; From each of the sixth local features, a seventh local feature is determined; the seventh local feature does not cover each of the sixth local features; and by combining the seventh local feature, the feature content of the ultrasonic data to be analyzed is obtained.

6. The method as described in claim 1, characterized in that, The step of obtaining attribute characterization data corresponding to each of the aforementioned feature contents includes: The real-time ultrasonic mapping content corresponding to each of the aforementioned features is obtained; the real-time ultrasonic mapping content corresponding to the feature content is obtained by matching the feature content through a real-time data matching unit. Each attribute characterization data corresponding to each of the aforementioned feature contents is obtained by using the real-time ultrasonic depiction content corresponding to each of the aforementioned feature contents.

7. The method as described in claim 6, characterized in that, The process of obtaining real-time ultrasonic mapping content corresponding to each of the aforementioned features includes: The database is used to search for candidate feature content corresponding to each of the aforementioned feature contents; the database records the association between candidate feature content and candidate matching information, and the candidate matching information is obtained by the real-time data matching unit matching the corresponding candidate feature content; When a candidate feature content corresponding to the feature content is found, the real-time ultrasonic depiction content corresponding to the feature content is obtained based on the candidate matching information matched by the found candidate feature content. If no candidate feature content corresponding to the feature content is found, the real-time data matching unit is invoked to match the feature content and obtain the real-time ultrasonic mapping content corresponding to the feature content.

8. The method as described in claim 1, characterized in that, The step of determining the first common factor between each feature content and each candidate category using the attribute characterization data and the category description data of each candidate category includes: determining the third common factor between each attribute characterization data and each candidate category using the attribute characterization data and the category label of each candidate category; and determining the first common factor between each feature content and each candidate category using the third common factor. The step of determining the third common factor between each attribute characterization data and each candidate category, using the attribute characterization data and the category label of each candidate category, includes: By using the attribute characterization data and the category labels of the candidate categories, the intersection features between the attribute characterization data and the category labels of the candidate categories are determined. From the intersection features of each attribute characterization data with the category label of each candidate category, determine the target intersection features of each attribute characterization data with the category label of each candidate category; The third proportion of the total number of features of the target intersection features between each of the attribute characterization data and the category label of each of the candidate categories to the total number of features of the category label of each of the candidate categories is determined; The third commonality factor between each attribute characterization data and the category label of each candidate species is determined by the third proportion, the first feature point of the target intersection feature between each attribute characterization data and the category label of each candidate species in the corresponding attribute characterization data, and the first abnormal text probability of the target intersection feature between each attribute characterization data and the category label of each candidate species; wherein, the target intersection feature between the attribute characterization data and the category label of the candidate species is not covered in the intersection features of the attribute characterization data and the category label of the candidate species other than itself. This also includes: determining a fourth common factor between each attribute characterization data and each candidate type by using the attribute characterization data, the pre-set type association features of each candidate type, and the pre-set type association features of each candidate type and the pre-set related parameters of the corresponding candidate type; determining a first common factor between each feature content and each candidate type by using the third common factor includes: determining a first common factor between each feature content and each candidate type by using the third common factor and the fourth common factor. The step of determining the fourth common factor between each attribute characterization data and each candidate type by means of each attribute characterization data, the pre-set type association features of each candidate type, and the pre-set type association features of each candidate type and the pre-set related parameters of the corresponding candidate type includes: determining the fourth common factor between each attribute characterization data and each candidate type by means of the second feature points of the pre-set type association features of each candidate type in each attribute characterization data, the second abnormal text probability of the pre-set type association features of each candidate type, and the pre-set related parameters of the pre-set type association features of each candidate type and the corresponding candidate type.

9. A data optimization system for an ultrasonic flow meter, characterized in that, The device includes a processor and a memory that communicate with each other, the processor being configured to read a computer program from the memory and execute it to implement the ultrasonic data optimization method according to any one of claims 1-8.

Images