Meter selection judgment and intelligent recommendation method based on statistical algorithm

Abstract

The present application relates to a method for meter selection determination and intelligent recommendation based on a statistical algorithm. The method includes collecting a first historical data set of the user's initial meter, the data set contains multiple records, each record includes at least an instantaneous flow value and a first cumulative flow value; based on the first cumulative flow value analysis, invalid records are determined and deleted from the data set to form a second historical data set; frequency statistics are performed on the instantaneous flow values ​​within the target time period, and the instantaneous flow value interval is determined based on these frequencies; based on the instantaneous flow value interval, the predicted upper and / or lower limits of the instantaneous flow values ​​of the user's initial meter in the target time period are determined to determine whether the meter selection is abnormal; if the meter selection is determined to be abnormal, a suitable target meter is recommended from the expert knowledge base to match the user's needs. This method can accurately and efficiently determine whether the user's meter selection is abnormal and make intelligent meter recommendations.

Description

Technical Field

[0001] The present application relates to the field of intelligent metering and data analysis, and in particular to a meter selection determination and intelligent recommendation method and computer equipment based on statistical algorithms. Background Art

[0002] In the process of selecting flow meters in the early stage, water supply enterprises are subject to the influence of professional technology and lack of data as a benchmark reference. The installation environment requirements for various flow meters are not clear, the user's water consumption and flow range are not accurately verified, and the user's water consumption needs and water use characteristics cannot be fully understood. It is impossible to clarify short-term and long-term planning, or the flow meter is reasonable when installed, but with the changes in the external environment, a large number of factors that are not conducive to the operation of the flow meter are derived (small water volume due to pressure reduction, low flow rate or unsuitable installation environment of the flow meter, etc.), resulting in a large number of abnormal problems of "big horse pulling a small cart" or "small horse pulling a big cart" with the flow meter, and ultimately causing the flow meter to measure too much, too little or unable to measure; therefore, whether the flow meter is reasonable in installation and adaptation will be crucial.

[0003] Scientific and reasonable water meter selection can achieve good results in improving water meter rationality, reducing water meter operation failures and maintenance costs, reducing metering losses, and achieving accurate and efficient metering. However, in related technologies, the rationality of the selection is determined by regularly analyzing the usage of the meter. However, due to complex factors such as flow fluctuations and peak usage periods, there are certain misjudgments in the determination of whether the selection is rational, and a lot of time is required for analysis, resulting in low analysis efficiency.

[0004] Therefore, the present invention application aims to accurately and efficiently determine whether the user's meter selection is abnormal and to make intelligent recommendations for meters. Summary of the invention

[0005] Based on this, it is necessary to provide a meter selection determination and intelligent recommendation method and computer equipment based on statistical algorithms that can accurately and efficiently determine whether the user's meter selection is abnormal and make intelligent recommendations for meters in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for determining and intelligently recommending meter selection based on a statistical algorithm. The method comprises:

[0007] Acquire a first historical data set of the user's initial meter; wherein the first historical data set includes a plurality of records, each record including at least an instantaneous flow value and a first cumulative flow value;

[0008] For the multiple records of the first historical data set, determine invalid records in the multiple records based at least on the first cumulative flow value in the multiple records, delete the invalid records in the multiple records, and obtain a second historical data set;

[0009] Based on the second historical data set, determine the instantaneous flow value of each record within the target time period, count the frequency of each instantaneous flow value within the target time period, and determine the instantaneous flow value interval according to each instantaneous flow value and the frequency corresponding to each instantaneous flow value;

[0010] Determine, according to at least part of the instantaneous flow values ​​in the instantaneous flow value interval, an upper limit of the instantaneous flow value prediction and / or a lower limit of the instantaneous flow value prediction of the user's initial meter in the target time period, and determine whether the user's initial meter selection is abnormal based on the upper limit of the instantaneous flow value prediction and / or the lower limit of the instantaneous flow value prediction of the user's initial meter;

[0011] If the user's initial meter selection is abnormal, obtain the expert knowledge base, and recommend a target meter that matches the user from the target meters in the expert knowledge base based on at least the instantaneous flow value prediction upper limit and / or instantaneous flow value prediction lower limit of the user's initial meter.

[0012] In one embodiment, determining the instantaneous flow value interval according to each instantaneous flow value and the frequency corresponding to each instantaneous flow value includes:

[0013] According to the size of each frequency in the frequency set, N frequencies are selected from the frequency set, and the instantaneous flow value interval is determined according to the instantaneous flow values ​​corresponding to the N frequencies; wherein the frequency set is composed of the frequencies corresponding to each instantaneous flow value.

[0014] In one embodiment, according to the magnitude of each frequency in the frequency set, N frequencies are selected from the frequency set, and the instantaneous flow value interval is determined according to the instantaneous flow values ​​corresponding to the N frequencies, including:

[0015] Sort the frequencies in the frequency set from large to small;

[0016] Generate an instantaneous flow value interval based on the instantaneous flow values ​​corresponding to the top N frequencies; wherein the ratio of the sum of the top N frequencies to the total number of records in the target time period is greater than the preset proportion.

[0017] In one embodiment, determining whether the user's initial meter selection is abnormal based on the instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit of the user's initial meter includes:

[0018] Obtaining the instantaneous flow value standard upper limit and / or instantaneous flow value standard lower limit of the user's initial meter;

[0019] Compare the instantaneous flow value prediction upper limit and the instantaneous flow value standard upper limit to determine whether a first determination interval exists, and / or compare the instantaneous flow value prediction lower limit and the instantaneous flow value standard lower limit to determine whether a second determination interval exists;

[0020] If the first determination interval and / or the second determination interval exist, it is determined that the user's initial meter selection is abnormal.

[0021] In one of the embodiments, comparing the instantaneous flow value prediction upper limit and the instantaneous flow value standard upper limit to determine whether there is a first determination interval includes: if the instantaneous flow value prediction upper limit is greater than the instantaneous flow value standard upper limit, then there is a first determination interval, and the first determination interval is: greater than the instantaneous flow value standard upper limit and less than the set upper limit threshold; comparing the instantaneous flow value prediction lower limit and the instantaneous flow value standard lower limit to determine whether there is a second determination interval includes: if the instantaneous flow value prediction lower limit is less than the instantaneous flow value standard lower limit, then there is a second determination interval, and the second determination interval is: less than the instantaneous flow value standard lower limit and greater than the set lower limit threshold.

[0022] In one embodiment, if the first determination interval and / or the second determination interval exist, determining that the user's initial meter selection is abnormal includes:

[0023] If there is a first determination interval, determine the number of consecutive target time periods that meet the following conditions: the ratio of the number of records of instantaneous flow values ​​in the first determination interval within the target time period to the total number of records within the target time period is greater than a set first threshold;

[0024] If the number of consecutive target time periods is greater than a set first number threshold, it is determined that the user's initial meter selection is abnormal;

[0025] If there is a second determination interval, determine the number of consecutive target time periods that meet the following conditions: the ratio of the number of records of instantaneous flow values ​​in the second determination interval within the target time period to the total number of records within the target time period is greater than a set second threshold;

[0026] If the number of consecutive target time periods is greater than a set second number threshold, it is determined that the user's initial meter selection is abnormal.

[0027] In one embodiment, the expert knowledge base includes first source data and second source data, and obtaining the expert knowledge base includes:

[0028] Acquire first user information and first instrument information corresponding to the first user information, wherein the instrument information recommended based on the information reported by the user is defined as the first instrument information, and the information reported by the user is defined as the first user information;

[0029] If the meter recommended based on the first user information does not have a selection anomaly within the first preset time period, the first user information and the first meter information corresponding to the first user information are used as the first data source;

[0030] Acquire second meter information and second user information; wherein the second meter information is meter information without selection abnormality within a second preset time period, and the second user information is user information corresponding to the second meter information, and the second meter information and the second user information are used as the second data source;

[0031] At least based on specific information in the first user information and the second user information, cluster analysis is performed on the first data source and the second data source to generate an expert knowledge base.

[0032] In one embodiment, the target meter that can match the user is recommended from the target meters in the expert knowledge base according to the instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit of the user's initial meter, including:

[0033] The upper limit of the instantaneous flow value prediction of the user's initial meter is expanded, and the lower limit of the instantaneous flow value prediction of the user's initial meter is expanded to obtain the expanded flow range;

[0034] At least based on the expanded flow range and the specific information of the user, a target meter that can match the user is recommended from the target meters in the expert knowledge base.

[0035] In one embodiment, for multiple records of the first historical data set, determining invalid records in the multiple records based at least on the first cumulative flow value in the multiple records, deleting the invalid records in the multiple records, and obtaining the second historical data set includes:

[0036] For multiple records of the first historical data set, determine a second cumulative flow value in each record; wherein the second cumulative flow value is obtained by performing a differential calculation on two first cumulative flow values ​​in two adjacent records;

[0037] Records in which the second cumulative flow value is less than the set flow are recorded as invalid records, and the invalid records in the plurality of records are deleted to obtain a second historical data set.

[0038] In one embodiment, each record in the first historical data set also includes environmental parameters, which include temperature and pressure. In addition to recording the records in which the second cumulative flow value is less than the set flow as invalid records, the method also records the records in which the temperature exceeds the temperature threshold and / or the pressure exceeds the pressure threshold as invalid records.

[0039] In a second aspect, the present application further provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a meter selection determination and intelligent recommendation method based on a statistical algorithm according to the first aspect is implemented.

[0040] The above-mentioned meter selection determination and intelligent recommendation method and computer device based on statistical algorithm collect the first historical data set of the user's initial meter, the first historical data set contains multiple records, each record includes at least instantaneous flow value and first cumulative flow value; based on the first cumulative flow value analysis, invalid records are determined and deleted from the first historical data set to form a second historical data set; based on the second historical data set, the instantaneous flow value in the target time period is counted, and the instantaneous flow value interval is determined according to these frequencies; based on the instantaneous flow value interval, the predicted upper and / or lower limits of the instantaneous flow value of the user's initial meter in the target time period are determined, which is used as the basis for determining whether the meter selection is abnormal; if the meter selection is determined to be abnormal, a suitable target meter is recommended from the expert knowledge base according to the instantaneous flow value predicted upper and / or instantaneous flow value predicted lower limits of the user's initial meter to match the user's needs. This method accurately analyzes the historical data of the user's initial meter, effectively identifies and excludes invalid records, accurately predicts the flow interval, and then accurately and quickly determines whether the meter selection is reasonable, and intelligently recommends suitable meters to improve the selection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a flow chart of a meter selection determination and intelligent recommendation method based on a statistical algorithm in one embodiment;

[0042] Figure 2 It is a schematic diagram of a flow chart of determining an instantaneous flow value interval according to an instantaneous flow value corresponding to a frequency in a frequency set in one embodiment;

[0043] Figure 3 It is a flowchart of determining whether the user's initial meter selection is abnormal based on the predicted upper limit and / or lower limit of the instantaneous flow value in one embodiment;

[0044] Figure 4 It is a schematic diagram of a flow chart of determining whether the first and second determination intervals exist in one embodiment;

[0045] Figure 5 A schematic diagram of a flow chart for determining whether the user's initial meter selection is abnormal if there is a first determination interval in an embodiment;

[0046] Figure 6 A schematic diagram of a flow chart for determining whether the user's initial meter selection is abnormal if there is a second determination interval in an embodiment;

[0047] Figure 7 A schematic diagram of a process for obtaining an expert database in one embodiment;

[0048] Figure 8 A schematic diagram of a process for recommending a matching target meter according to an upper limit and / or a lower limit of an instantaneous flow value prediction in one embodiment;

[0049] Fig. 9 A schematic diagram of a process for determining a second historical data set in one embodiment;

[0050] Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] In one embodiment, Figure 1 As shown, a method for selecting and intelligently recommending a meter based on a statistical algorithm is provided, comprising the following steps:

[0053] Step 101: Acquire a first historical data set of an initial meter of a user; wherein the first historical data set includes a plurality of records, each record including at least an instantaneous flow value and a first cumulative flow value;

[0054] Specifically, the historical data of the user's initial meter is collected to form a first historical data set. The first historical data set may be composed of historical data of flow, temperature and pressure from the time when the user installs the meter to the time when the data is recorded, which may be 2, 3 or 5 years. The first historical data set consists of multiple records, each record including an instantaneous flow value and a first cumulative flow value at a certain point in time or time period. It should be noted that, in addition to the instantaneous flow value and the first cumulative flow value at a certain point in time or time period, each record may also include environmental data such as pressure or other flow values ​​when the initial meter is running. It should be further explained that the instantaneous flow value in each record may be the flow measured by the meter at a specific point in time or within a short period of time, and the first cumulative flow value may be the total flow from the installation of the meter to the time when the data is recorded.

[0055] Step 102: for the multiple records of the first historical data set, determine invalid records in the multiple records based on at least the first cumulative flow value in the multiple records, delete the invalid records in the multiple records, and obtain a second historical data set;

[0056] Specifically, the first cumulative flow value is the total flow from the installation of the meter to the time when the data is recorded. By analyzing this value, records with abnormal cumulative flow values ​​can be identified. If the cumulative flow value of a record is too low or too high, much lower or higher than adjacent records, it may indicate that the data in the record is abnormal. It should be noted that if the data in the record is abnormal, a large error will occur in the process of calculating the instantaneous flow value and the first cumulative flow value, affecting the result, so the record containing abnormal data is regarded as an invalid record. Once an invalid record is identified, the invalid record will be deleted from the first historical data set. After deleting the invalid records, the remaining records constitute the second historical data set.

[0057] Step 103: Based on the second historical data set, determine the instantaneous flow value of each record within the target time period, count the frequency of each instantaneous flow value within the target time period, and determine the instantaneous flow value interval according to each instantaneous flow value and the frequency corresponding to each instantaneous flow value;

[0058] Specifically, all recorded instantaneous flow values ​​within a target time period are extracted from the second historical data set. The target time period can be any set time period, such as one week, one month, or two months. For all instantaneous flow values ​​within the target time period, the number of occurrences of each flow value, i.e., the frequency, is counted. According to each instantaneous flow value and its corresponding frequency, the distribution interval of the instantaneous flow value, i.e., the instantaneous flow value interval, can be determined.

[0059] Step 104: determining an upper limit of the instantaneous flow value prediction and / or a lower limit of the instantaneous flow value prediction of the user's initial meter in the target time period according to at least part of the instantaneous flow value in the instantaneous flow value interval, and determining whether the user's initial meter selection is abnormal based on the upper limit of the instantaneous flow value prediction and / or the lower limit of the instantaneous flow value prediction of the user's initial meter;

[0060] Specifically, in step 103, the distribution interval of the instantaneous flow value has been determined. The instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit are determined using at least part of the instantaneous flow values ​​in the distribution interval of the instantaneous flow value. The instantaneous flow value prediction upper limit may refer to the maximum instantaneous flow value that occurs within the target time period, and the instantaneous flow value prediction lower limit may refer to the minimum instantaneous flow value that may occur. These values ​​can be determined by statistical analysis methods, such as calculating the percentiles of the flow values ​​in the interval (for example, the 95% percentile as the upper limit and the 5% percentile as the lower limit), or according to the shape of the flow distribution (such as a normal distribution). If the instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit are determined, these values ​​can be compared with the design flow range of the user's initial meter. If the instantaneous flow value prediction upper limit exceeds the design upper limit of the meter, or the instantaneous flow value prediction lower limit is lower than the design lower limit of the meter, it can be determined that the meter selection is abnormal, that is, the flow range of the meter is not suitable for the actual use needs of the user.

[0061] Step 105: If the user's initial meter selection is abnormal, obtain an expert knowledge base, and recommend a target meter that matches the user from the target meters in the expert knowledge base based on at least the instantaneous flow value prediction upper limit and / or instantaneous flow value prediction lower limit of the user's initial meter.

[0062] Specifically, in step 104, if it is determined that the user's initial meter selection is abnormal, that is, the flow range of the meter is not suitable for the user's actual usage needs. The expert knowledge base is a database containing a large amount of meter selection experience and historical data. The expert knowledge base may include meter selection cases for different users and different usage scenarios, as well as corresponding flow parameters and environmental conditions, such as temperature, pressure, etc. The upper and / or lower limits of the instantaneous flow value prediction of the user's initial meter are used to reflect the user's actual flow demand. Based on the user's actual flow demand and the information in the expert knowledge base, one or more target meters that can meet these demands are recommended.

[0063] In this embodiment, the first historical data set of the user's initial meter is collected, and these data sets contain multiple records, each record includes at least an instantaneous flow value and a first cumulative flow value; invalid records are determined based on the first cumulative flow value analysis and deleted from the data set to form a second historical data set; the instantaneous flow values ​​in the target time period are counted for frequency, and the instantaneous flow value interval is determined based on these frequencies; the predicted upper and / or lower limits of the instantaneous flow values ​​of the user's initial meter in the target time period are determined based on the instantaneous flow value interval, and this is used as a basis for determining whether the meter selection is abnormal; if the meter selection is determined to be abnormal, a suitable target meter is recommended from the expert knowledge base to match the user's needs. This method accurately analyzes the historical data of the user's initial meter, effectively identifies and excludes invalid records, accurately predicts the flow interval, quickly determines whether the meter selection is reasonable, and intelligently recommends suitable meters to improve selection accuracy.

[0064] In one embodiment, determining the instantaneous flow value interval according to each instantaneous flow value and the frequency corresponding to each instantaneous flow value includes:

[0065] According to the size of each frequency in the frequency set, N frequencies are selected from the frequency set, and the instantaneous flow value interval is determined according to the instantaneous flow values ​​corresponding to the N frequencies; wherein the frequency set is composed of the frequencies corresponding to each instantaneous flow value.

[0066] Specifically, for each different instantaneous flow value, the number of times it appears in the target time period, i.e., the frequency, is calculated. The frequencies corresponding to all different instantaneous flow values ​​together constitute a frequency set. N frequencies are directly selected from the frequency set, where N can be a preset number that can represent the number of frequencies that are considered most important in the instantaneous flow value analysis. After determining the N frequencies, the instantaneous flow values ​​corresponding to these frequencies are found, and based on these instantaneous flow values, one or more instantaneous flow value intervals can be determined.

[0067] In this embodiment, a frequency set is constructed based on the frequencies corresponding to each instantaneous flow value, and the set contains the frequency data of all instantaneous flow values. N frequencies are selected from this frequency set to ensure that the selected frequencies can represent the main flow changes in the data set. The instantaneous flow value interval is determined based on the instantaneous flow values ​​corresponding to the N frequencies. Such interval division can more accurately reflect the actual usage of the user's meter. This method makes the upper and lower limits of the instantaneous flow value prediction more accurate through refined frequency analysis and interval division, which helps to more accurately determine whether the meter selection is abnormal.

[0068] In one embodiment, Figure 2 As shown, according to the size of each frequency in the frequency set, N frequencies are selected from the frequency set, and the instantaneous flow value interval is determined according to the instantaneous flow values ​​corresponding to the N frequencies, including the following steps:

[0069] Step 201: sort the frequencies in the frequency set from large to small;

[0070] Specifically, the frequencies in the frequency set are sorted from large to small. After sorting, the data in the frequency set will be arranged in descending order of frequency.

[0071] Step 202: Generate an instantaneous flow value interval according to each instantaneous flow value corresponding to the top N frequencies; wherein the ratio of the sum of the top N frequencies to the total number of records in the target time period is greater than a preset ratio.

[0072] Specifically, one or more instantaneous flow value intervals are generated based on the instantaneous flow values ​​corresponding to the top N frequencies. The ratio of the sum of the top N frequencies to the total number of records in the target time period needs to be greater than a preset proportion. The preset proportion is a threshold used to ensure that the selected instantaneous flow value interval can represent enough data records to be statistically significant. For example, if the preset proportion is 80%, the number of records covered by the top N frequencies accounts for at least 80% of the total number of records.

[0073] In this embodiment, the frequencies in the frequency set are sorted from large to small, and the top N frequencies are selected. The instantaneous flow values ​​corresponding to these frequencies are used to generate instantaneous flow value intervals. This process ensures that the ratio of the sum of the top N frequencies to the total number of records in the target time period is greater than a preset proportion, thereby ensuring that the selected frequencies are representative and can reflect the flow changes in most cases. This method determines the flow interval by selecting the instantaneous flow values ​​corresponding to the high-frequency numbers, thereby improving the accuracy of meter selection and the applicability of intelligent recommendations.

[0074] In one embodiment, Figure 3 As shown, judging whether the user's initial meter selection is abnormal based on the instantaneous flow value prediction upper limit and / or instantaneous flow value prediction lower limit of the user's initial meter includes the following steps:

[0075] Step 301: Obtaining the instantaneous flow value standard upper limit and / or the instantaneous flow value standard lower limit of the user's initial meter;

[0076] Specifically, it is necessary to obtain the standard upper limit and / or standard lower limit of the instantaneous flow value of the initial meter currently used by the user. These values ​​are usually provided by the meter manufacturer or determined based on industry standards and specifications. The standard upper limit of the instantaneous flow value refers to the maximum instantaneous flow value that the meter is designed to accurately measure or the maximum instantaneous flow value measured within the allowable range. Exceeding the standard upper limit of the instantaneous flow value may result in inaccurate measurement or damage to the meter. The standard lower limit of the instantaneous flow value refers to the minimum instantaneous flow value that the meter can accurately measure or the minimum instantaneous flow value measured within the allowable range. The flow below the standard lower limit of the instantaneous flow value may not be accurately measured.

[0077] Step 302: comparing the instantaneous flow value prediction upper limit and the instantaneous flow value standard upper limit to determine whether a first determination interval exists, and / or comparing the instantaneous flow value prediction lower limit and the instantaneous flow value standard lower limit to determine whether a second determination interval exists;

[0078] Specifically, the predicted upper limit of the instantaneous flow value of the user's initial meter is compared with the standard upper limit. If the predicted upper limit is greater than the standard upper limit, that is, at certain moments, the flow of the meter may exceed its design range, which may indicate that the meter is too small. If the predicted upper limit is greater than the standard upper limit, then a first determination interval can be determined, which is a range greater than the standard upper limit and less than the set lower limit threshold. The existence of the first determination interval indicates that the meter may not be able to accurately measure the flow within the interval or is outside the range of the maximum instantaneous flow value allowed to be measured, and it is necessary to consider replacing a meter with a larger range.

[0079] Compare the predicted lower limit of the instantaneous flow value of the user's initial meter with the standard lower limit. If the predicted lower limit is less than the standard lower limit, that is, at certain moments, the flow of the meter may be lower than its design range, which may indicate that the meter selection is too large. If the predicted lower limit is less than the standard lower limit, then a second judgment interval can be determined. This interval is a range that is less than the standard lower limit and greater than the set lower limit threshold. The existence of the second judgment interval indicates that the meter may not be able to accurately measure the flow within this interval or is outside the range of the minimum instantaneous flow value allowed to be measured, and it is necessary to consider replacing a meter with a smaller range.

[0080] Step 303: If the first determination interval and / or the second determination interval exist, it is determined that the user's initial meter selection is abnormal.

[0081] Specifically, if there is a first determination interval, that is, the upper limit of the predicted instantaneous flow value is greater than the standard upper limit of the meter design and less than the set lower limit threshold, this indicates that in some time periods, the actual flow exceeds the maximum value that the meter can accurately measure. If there is a second determination interval, that is, the lower limit of the predicted instantaneous flow value is less than the standard lower limit of the meter design and greater than the set lower limit threshold, this indicates that in some time periods, the actual flow is lower than the minimum value that the meter can accurately measure. If any one of the determination intervals exists, that is, the first determination interval or the second determination interval, or both, then the user's initial meter selection will be judged as abnormal.

[0082] In this embodiment, the standard upper limit and / or standard lower limit of the instantaneous flow value of the user's initial meter is obtained; the predicted upper limit of the instantaneous flow value is compared with the standard upper limit, and the predicted lower limit of the instantaneous flow value is compared with the standard lower limit to determine whether there is a first determination interval and / or a second determination interval; if these determination intervals exist, the user's initial meter selection is determined to be abnormal. This method provides an accurate determination mechanism that can identify the irrationality of meter selection by comparing the predicted value and the standard value, so as to make timely adjustments or replacements to ensure that the meter selection matches the actual usage needs of the user.

[0083] In one embodiment, Figure 4 As shown, the following steps are included:

[0084] Step 401: comparing the instantaneous flow value prediction upper limit and the instantaneous flow value standard upper limit to determine whether there is a first determination interval, including: if the instantaneous flow value prediction upper limit is greater than the instantaneous flow value standard upper limit, then there is a first determination interval, and the first determination interval is: greater than the instantaneous flow value standard upper limit and less than the set upper limit threshold;

[0085] Specifically, the predicted upper limit of the instantaneous flow value of the user's initial meter is compared with the standard upper limit of the meter. The predicted upper limit of the instantaneous flow value is based on the user's historical data and statistical analysis, while the standard upper limit is the maximum measurement value determined by the manufacturer when the meter is designed. If the predicted upper limit of the instantaneous flow value is greater than the standard upper limit of the instantaneous flow value, that is, during certain time periods, the user's actual flow demand may exceed the designed measurement capacity of the meter. In this case, it can be determined that there is a first determination interval, and the first determination interval is defined as greater than the standard upper limit of the instantaneous flow value and less than the set upper limit threshold. It should be noted that the upper limit threshold can be set according to an empirical value, or it can be the predicted upper limit of the instantaneous flow value.

[0086] Step 402: Compare the predicted lower limit of the instantaneous flow value and the standard lower limit of the instantaneous flow value to determine whether there is a second determination interval, including: if the predicted lower limit of the instantaneous flow value is less than the standard lower limit of the instantaneous flow value, then there is a second determination interval, and the second determination interval is: less than the standard lower limit of the instantaneous flow value and greater than the set lower limit threshold.

[0087] Specifically, the predicted lower limit of the instantaneous flow value of the user's initial meter is compared with the standard lower limit of the meter. The predicted lower limit of the instantaneous flow value is based on the user's historical data and statistical analysis, while the standard lower limit is the minimum measurement value determined by the manufacturer when the meter is designed. If the predicted lower limit of the instantaneous flow value is less than the standard lower limit of the instantaneous flow value, that is, in certain time periods, the user's actual flow demand may be lower than the designed measurement sensitivity of the meter. In this case, it can be determined that there is a second determination interval, which is less than the standard lower limit of the instantaneous flow value and greater than the set lower limit threshold, where the lower limit threshold can be set according to an empirical value, or it can be the predicted lower limit of the instantaneous flow value.

[0088] In this embodiment, if the predicted upper limit of the instantaneous flow value is greater than the standard upper limit of the instantaneous flow value, it is determined that there is a first determination interval, which is defined as greater than the standard upper limit of the instantaneous flow value and less than the set upper limit threshold; if the predicted lower limit of the instantaneous flow value is less than the standard lower limit of the instantaneous flow value, it is determined that there is a second determination interval, which is defined as less than the standard lower limit of the instantaneous flow value and greater than the set lower limit threshold. This method provides a clear determination logic, which enables accurate identification of situations where the meter selection does not meet the actual use requirements, and timely adjustment or replacement to ensure that the meter selection is reasonable.

[0089] In one embodiment, Figure 5 As shown, if there is a first determination interval, it is determined that the user's initial meter selection is abnormal, including the following steps:

[0090] Step 501: if there is a first determination interval, determine the number of consecutive target time periods that meet the following conditions: the ratio of the number of records of instantaneous flow values ​​in the first determination interval within the target time period to the total number of records within the target time period is greater than a set first threshold;

[0091] Specifically, the first determination interval is a range greater than the standard upper limit of the instantaneous flow value and less than the set prediction upper limit. The first determination interval may be a flow value that represents a flow value that exceeds the design measurement range of the meter. It is necessary to determine the number of continuous target time periods in which the ratio of the number of instantaneous flow value records to the total number of records within the first determination interval is greater than the set first threshold. That is, in these continuous target time periods, a considerable portion of the flow records exceeds the standard upper limit of the meter. It should be noted that one month can be a target time period. Among them, the first threshold is a preset ratio used to judge the degree of abnormality of the instantaneous flow value record within the continuous target time period. For example, if the first threshold is set to 80%, the target time period can be a week or a month, then in each consecutive week or month, at least 80% of the flow records exceed the standard upper limit of the meter.

[0092] Step 502: If the number of consecutive target time periods is greater than a set first number threshold, it is determined that the user's initial meter selection is abnormal.

[0093] Specifically, the first quantity threshold is a quantity standard set in advance, which is used to determine whether the situation where the flow exceeds the upper limit of the meter standard occurs frequently. The number of continuous target time periods that meet the conditions is compared with a preset first quantity threshold. If the number of continuous target time periods that meet the conditions exceeds this preset first quantity threshold, then the user's initial meter selection will be judged to be abnormal. For example, the target time period can be one month. If the number of continuous target time periods that meet the conditions is greater than the set first quantity threshold, that is, the number of continuous months that meet the conditions is greater than the set first quantity threshold, then it can be judged that the user's initial meter selection is abnormal.

[0094] In one embodiment, Figure 6 As shown, if there is a second determination interval, it is determined that the user's initial meter selection is abnormal, including the following steps:

[0095] Step 601: if there is a second determination interval, determine the number of consecutive target time periods that meet the following conditions: the ratio of the number of records of instantaneous flow values ​​in the second determination interval within the target time period to the total number of records within the target time period is greater than a set second threshold;

[0096] Specifically, the second determination interval is a range that is less than the standard lower limit of the instantaneous flow value and greater than the set lower limit threshold. This interval represents those flow values ​​that are lower than the design measurement range of the meter. It is necessary to determine the number of continuous target time periods in which the ratio of the number of instantaneous flow value records to the total number of records in the second determination interval is greater than the set second threshold. That is, in these continuous target time periods, a considerable portion of the flow records are lower than the standard lower limit of the meter. It should be noted that one month can be a target time period. Among them, the second threshold is a preset ratio used to judge the degree of abnormality of the instantaneous flow value records in the continuous target time period. For example, if the second threshold is set to 80%, the target time period can be one week or one month, then in each consecutive week or month, at least 80% of the flow records are lower than the standard lower limit of the meter.

[0097] Step 602: If the number of consecutive target time periods is greater than a set second number threshold, it is determined that the user's initial meter selection is abnormal.

[0098] Specifically, the second quantity threshold is a quantity standard set in advance, which is used to determine whether the situation where the flow rate is lower than the lower limit of the meter standard occurs frequently. The number of continuous target time periods that meet the conditions is compared with a preset second quantity threshold. If the number of continuous target time periods that meet the conditions exceeds this preset second quantity threshold, then the user's initial meter selection will be judged to be abnormal. For example, the target time period can be one month. If the number of continuous target time periods that meet the conditions is greater than the set second quantity threshold, that is, the number of continuous months that meet the conditions is greater than the set second quantity threshold, then it can be judged that the user's initial meter selection is abnormal.

[0099] The specific judgment can be expressed in the following table:

[0100]

[0101] In this embodiment, if there is a first determination interval, that is, the instantaneous flow value prediction upper limit is greater than the standard upper limit and less than the set upper limit threshold, then the ratio of the number of records in which the instantaneous flow value falls within this interval to the total number of records in the target time period will be calculated. If the ratio is greater than the set first threshold, and the number of target time periods that continuously meet this condition exceeds the set first number threshold, then the meter selection is determined to be abnormal. Similarly, if there is a second determination interval, that is, the instantaneous flow value prediction lower limit is less than the standard lower limit and greater than the set lower limit threshold, then the ratio of the number of records in which the instantaneous flow value falls within this interval to the total number of records in the target time period will be calculated. If the ratio is greater than the set second threshold, and the number of target time periods that continuously meet this condition exceeds the set second number threshold, then the meter selection is also determined to be abnormal. This method improves the accuracy and reliability of the determination by introducing the concepts of continuity and threshold, and can avoid erroneously determining that the meter selection is abnormal due to accidental flow anomalies, ensuring that determination is made only when flow anomalies continue to occur, thereby reducing the possibility of misjudgment.

[0102] In one embodiment, Figure 7 As shown, the expert knowledge base includes first source data and second source data, and obtaining the expert database includes the following steps:

[0103] Step 701: obtaining first user information and first instrument information corresponding to the first user information, wherein the instrument information recommended based on the information reported by the user is defined as the first instrument information, and the information reported by the user is defined as the first user information;

[0104] Specifically, basic information of the user is collected, and the basic information of the user is referred to as the first user information. The basic information of the user may include the specific usage scenario of the user (such as home, industry, business, etc.), gas or water consumption, usage frequency, time period, and other specific needs related to gas or water consumption. The first user information refers to the data directly reported by the user to the service provider, which is essential for understanding the specific needs of the user. The first user information may include the user's historical gas or water consumption data, expected flow demand, etc. It also involves collecting meter information corresponding to the first user information, and the meter information corresponding to the first user information is referred to as the first meter information. The first meter information may include the meter type, specifications, performance parameters, etc. recommended based on the user reported information. The first meter information is the meter information recommended by experts based on the information reported by the user (the first user information). The meter information recommended by the expert may be based on expert experience, historical data analysis, or a specific selection algorithm. There is a direct correspondence between the first user information and the first meter information, that is, specific user characteristics and needs correspond to specific meter recommendations.

[0105] Step 702: If the meter recommended based on the first user information does not have a selection anomaly within a first preset time period, the first user information and the first meter information corresponding to the first user information are used as a first data source;

[0106] Specifically, the first preset time period is a predetermined time range used to observe and record whether the recommended meter meets the actual needs of the user. The first preset time period can be several months or a year, depending on the frequency of use of the meter and industry standards. During the first preset time period, monitor whether there are any selection anomalies in the meter, such as flow exceeding the measurement range of the meter, frequent failures or other performance issues. If there are no selection anomalies in the recommended meter within the first preset time period, it indicates that the recommendation is successful and the corresponding first user information and first meter information are reliable. The verified first user information and second meter information will be included in the expert knowledge base as the first data source.

[0107] Step 703: Acquire second meter information and second user information; wherein the second meter information is meter information without selection abnormality within a second preset time period, and the second user information is user information corresponding to the second meter information, and the second meter information and the second user information are used as the second data source;

[0108] Specifically, the meter information (second meter information) that performs stably and has no selection anomalies within the second preset time period, as well as the user information (second user information) corresponding to these meters, are collected. The second preset time period is a specific time range used to evaluate the long-term performance and stability of the meter. The second preset time period is usually longer than the first preset time period to ensure that the collected meter information has sufficient long-term stability. During the second preset time period, the performance of the meter will be continuously monitored to ensure that no selection anomalies occur, such as flow exceeding the measurement range, frequent failures or other performance issues. If no selection anomalies occur in the meter within the second preset time period, it indicates that the meter selection is successful and the corresponding second user information and second meter information are reliable. The second user information and second meter information that have been verified for a long time will be included in the expert knowledge base as a second data source.

[0109] Step 704: Perform cluster analysis on the first data source and the second data source based on at least specific information in the first user information and the second user information to generate an expert knowledge base.

[0110] Specifically, the first data source and the second data source are integrated. Specific key information is selected from the first user information and the second user information, which may include the user's industry type, gas or water consumption, frequency of use, time period, etc., as well as the corresponding meter type, specifications and performance parameters. Use statistical and data mining techniques, such as clustering algorithms, to perform cluster analysis on the integrated data. Cluster analysis can group data with similar characteristics to better understand data patterns and trends. The results of the cluster analysis will be used to generate an expert knowledge base, which will contain meter selection patterns and best practices for different user groups. The expert knowledge base will be used for intelligent recommendations to help identify matches between new user information and existing patterns in the knowledge base, thereby providing accurate meter selection recommendations for new users.

[0111] In this embodiment, first user information and corresponding first meter information are obtained, and these information are obtained by recommending meters based on information reported by the user. If these recommended meters have no selection anomalies within the first preset time period, then this information will be used as the first data source. Second meter information and corresponding second user information are obtained, and these information involve meters that have not had selection anomalies within the second preset time period and their corresponding user information, and these are used as the second data source. Based on specific information in the first user information and the second user information, these two types of data are clustered and analyzed to generate an expert knowledge base. This method builds a comprehensive and verified expert knowledge base by combining user-reported information and actual operating data, which can provide more accurate and reliable intelligent recommendations.

[0112] In one embodiment, Figure 8 As shown, at least according to the instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit of the user's initial meter, recommending a target meter that can match the user from the target meters in the expert knowledge base includes the following steps:

[0113] Step 801: expanding the upper limit of the instantaneous flow value prediction of the user's initial meter, and expanding the lower limit of the instantaneous flow value prediction of the user's initial meter, to obtain an expanded flow range;

[0114] Specifically, the upper limit of the instantaneous flow value prediction of the user's initial meter is expanded, that is, an additional value is added on the basis of the upper limit of the prediction to ensure that all possible high flow conditions are covered. The lower limit of the instantaneous flow value prediction of the user's initial meter is expanded, that is, an additional value is reduced on the basis of the lower limit of the prediction to ensure that all possible low flow conditions are covered. By expanding the upper limit and the lower limit, a wider flow range is obtained. This flow range will include all possible instantaneous flow values, thereby providing a more comprehensive reference for meter selection.

[0115] Step 802: Recommend a target meter that can match the user from the target meters in the expert knowledge base at least based on the expanded flow range and the user's specific information.

[0116] Specifically, first consider the expanded predicted upper and lower limits of the instantaneous flow value of the user's initial meter. This range covers the highest and lowest flow conditions that the user may encounter. In addition to the flow range, user-specific information also needs to be considered, which may include the user's industry type, geographic location, usage habits, environmental conditions (such as temperature, pressure), and other factors that may affect meter performance. Utilize the data in the expert knowledge base, which contains historical data, expert experience, long-term stable operation data, etc., to filter out suitable meter options based on the expanded flow range and user-specific information. Recommend one or more target meters that can meet the user's flow requirements and specific conditions.

[0117] In this embodiment, the upper limit of the instantaneous flow value prediction of the user's initial meter is expanded, and the lower limit of the instantaneous flow value prediction is expanded, so as to obtain a wider flow range. Based on this expanded flow range and the user's specific information, the target meter that can match the user's needs is screened and recommended from the expert knowledge base. By expanding the flow range, this method increases the flexibility and adaptability of the recommendation, and can take into account more possible flow changes, thereby improving the accuracy and practicality of the recommendation.

[0118] In one embodiment, Fig. 9 As shown, for multiple records of the first historical data set, determining invalid records in the multiple records based on at least the first cumulative flow value in the multiple records, deleting the invalid records in the multiple records, and obtaining a second historical data set includes the following steps:

[0119] Step 901: for a plurality of records of a first historical data set, determine a second cumulative flow value in each record; wherein the second cumulative flow value is obtained by performing a differential calculation on two first cumulative flow values ​​in two adjacent records;

[0120] Specifically, for each record in the first historical data set, its second cumulative flow value needs to be determined. This value is obtained by performing a differential calculation on the two first cumulative flow values ​​in two adjacent records. The differential calculation is completed by subtracting the first cumulative flow value of the previous record from the first cumulative flow value of the current record. The calculation process can be expressed as:

[0121] ;

[0122] in, Indicates the cumulative flow value of a record. Indicates the cumulative flow value of the previous record. It represents the second cumulative flow value, that is, the difference between the cumulative flow values ​​of two adjacent records, where the interval between the two adjacent records can be one hour or half an hour.

[0123] Step 902: Records in which the second cumulative flow value is less than the set flow are recorded as invalid records, and the invalid records among the multiple records are deleted to obtain a second historical data set.

[0124] Specifically, a flow threshold needs to be set. The flow threshold is based on the minimum measured flow of the meter or based on historical data analysis, and is used to determine whether the flow record is valid. For each record in the first historical data set, if its second cumulative flow value is less than the set flow threshold, the record will be marked as an invalid record. These invalid records may represent situations where the flow is too low and cannot be accurately measured by the meter. If an invalid record is identified, the invalid record will be deleted from the first historical data set to form a second historical data set.

[0125] In this embodiment, the second cumulative flow value is obtained by differential calculation of the two first cumulative flow values ​​in two adjacent records. If the second cumulative flow value is less than the set flow threshold, the record is marked as an invalid record. By deleting these invalid records, a more accurate and reliable second historical data set is obtained. This method improves the quality of data and the accuracy of analysis, and ensures that subsequent analysis is based on more real and representative data by excluding records of small flow values ​​caused by measurement errors or other abnormal conditions.

[0126] In one embodiment, each record in the first historical data set also includes environmental parameters, which include temperature and pressure. In addition to recording records in which the second cumulative flow value is less than the set flow as invalid records, this method also includes recording records in which the temperature exceeds a temperature threshold and / or the pressure exceeds a pressure threshold as invalid records.

[0127] Specifically, the first historical data set contains not only instantaneous flow values ​​and cumulative flow values, but also environmental parameters corresponding to each record, which mainly include temperature, pressure, humidity, etc. In addition to identifying invalid records based on flow values, this method also takes into account the impact of environmental parameters on data accuracy. If the temperature value in the record exceeds the preset temperature threshold, the pressure value exceeds the preset pressure threshold, and the humidity value exceeds the preset humidity threshold, these records may also be considered invalid. The temperature threshold, pressure threshold, and humidity threshold are based on the normal working conditions of the meter and the analysis of historical data. These preset thresholds define the environmental parameter range for which the meter measurement data is considered to be a valid record. If a record is marked as invalid, whether because the flow value or the environmental parameter exceeds the threshold, the record will be deleted from the first historical data set to form a more accurate and reliable second historical data set.

[0128] In this embodiment, in the process of determining invalid records, in addition to marking the records whose second cumulative flow value is less than the set flow threshold as invalid, the records whose temperature exceeds the set temperature threshold and / or the pressure exceeds the set pressure threshold are also marked as invalid. In this way, these invalid records can be deleted from the first historical data set to obtain a more accurate and reliable second historical data set. By comprehensively considering the flow and environmental parameters, this method can more comprehensively identify and exclude abnormal records that may affect the accuracy of data analysis.

[0129] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0130] Based on the same concept, the present application also provides a computer device. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for selecting and intelligently recommending a meter based on a statistical algorithm is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0131] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0132] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for selecting and intelligently recommending a meter based on a statistical algorithm, characterized in that: The method comprises: Acquire a first historical data set of the user's initial meter; wherein the first historical data set includes a plurality of records, each record including at least an instantaneous flow value and a first cumulative flow value; For the plurality of records in the first historical data set, determining invalid records in the plurality of records based at least on the first cumulative flow value in the plurality of records, deleting the invalid records in the plurality of records, and obtaining a second historical data set; Based on the second historical data set, determine the instantaneous flow value of each record within the target time period, count the frequency of each instantaneous flow value within the target time period, and determine the instantaneous flow value interval according to each instantaneous flow value and the frequency corresponding to each instantaneous flow value; Determine, according to at least part of the instantaneous flow values ​​in the instantaneous flow value interval, an upper limit of the instantaneous flow value prediction and / or a lower limit of the instantaneous flow value prediction of the user's initial meter in the target time period, and determine whether the selection of the user's initial meter is abnormal based on the upper limit of the instantaneous flow value prediction and / or the lower limit of the instantaneous flow value prediction of the user's initial meter; If the user's initial meter selection is abnormal, an expert knowledge base is obtained, and a target meter that can match the user is recommended from the target meters in the expert knowledge base at least based on the instantaneous flow value prediction upper limit and / or instantaneous flow value prediction lower limit of the user's initial meter.

2. The meter selection determination and intelligent recommendation method based on statistical algorithm according to claim 1 is characterized in that: Determining the instantaneous flow value interval according to each instantaneous flow value and the frequency corresponding to each instantaneous flow value includes: According to the size of each frequency in the frequency set, N frequencies are selected from the frequency set, and the instantaneous flow value interval is determined according to the instantaneous flow values ​​corresponding to the N frequencies; wherein the frequency set is composed of the frequencies corresponding to the various instantaneous flow values.

3. The meter selection determination and intelligent recommendation method based on statistical algorithm according to claim 2 is characterized in that: According to the size of each frequency in the frequency set, N frequencies are selected from the frequency set, and the instantaneous flow value interval is determined according to the instantaneous flow values ​​corresponding to the N frequencies, including: Sort the frequencies in the frequency set from large to small; An instantaneous flow value interval is generated according to each instantaneous flow value corresponding to the top N frequencies; wherein the ratio of the sum of the top N frequencies to the total number of records in the target time period is greater than a preset ratio.

4. The meter selection determination and intelligent recommendation method based on statistical algorithm according to any one of claims 1 to 3, characterized in that: Determining whether the user's initial meter selection is abnormal based on the instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit of the user's initial meter includes: Obtaining the instantaneous flow value standard upper limit and / or instantaneous flow value standard lower limit of the user's initial meter; Compare the instantaneous flow value prediction upper limit and the instantaneous flow value standard upper limit to determine whether a first determination interval exists, and / or compare the instantaneous flow value prediction lower limit and the instantaneous flow value standard lower limit to determine whether a second determination interval exists; If the first determination interval and / or the second determination interval exists, it is determined that the user's initial meter selection is abnormal.

5. The method for selecting and intelligently recommending a meter based on a statistical algorithm according to claim 4, characterized in that: Comparing the instantaneous flow value prediction upper limit and the instantaneous flow value standard upper limit to determine whether there is a first determination interval includes: if the instantaneous flow value prediction upper limit is greater than the instantaneous flow value standard upper limit, then there is a first determination interval, and the first determination interval is: greater than the instantaneous flow value standard upper limit and less than a set upper limit threshold; Comparing the instantaneous flow value prediction lower limit and the instantaneous flow value standard lower limit to determine whether a second determination interval exists includes: if the instantaneous flow value prediction lower limit is less than the instantaneous flow value standard lower limit, then there is a second determination interval, and the second determination interval is: less than the instantaneous flow value standard lower limit and greater than the set lower limit threshold.

6. The method for selecting and intelligently recommending a meter based on a statistical algorithm according to claim 5, characterized in that: If the first determination interval and / or the second determination interval exist, determining that the user's initial meter selection is abnormal includes: If the first determination interval exists, determine the number of consecutive target time periods that meet the following conditions: the ratio of the number of records of instantaneous flow values ​​in the first determination interval within the target time period to the total number of records within the target time period is greater than a set first threshold; if the number of consecutive target time periods is greater than the set first number threshold, determine that the user's initial meter selection is abnormal; If the second determination interval exists, determine the number of consecutive target time periods that meet the following conditions: the ratio of the number of records of instantaneous flow values ​​in the second determination interval within the target time period to the total number of records within the target time period is greater than a set second threshold; if the number of consecutive target time periods is greater than the set second number threshold, determine that the user's initial meter selection is abnormal.

7. The meter selection determination and intelligent recommendation method based on statistical algorithm according to any one of claims 1 to 3, characterized in that: The expert knowledge base includes first source data and second source data, and obtaining the expert database includes: Acquire first user information and first instrument information corresponding to the first user information, wherein the instrument information recommended based on the information reported by the user is defined as the first instrument information, and the information reported by the user is defined as the first user information; If the meter recommended based on the first user information does not have a selection abnormality within a first preset time period, the first user information and the first meter information corresponding to the first user information are used as a first data source; Acquire second meter information and second user information; wherein the second meter information is meter information without selection abnormality within a second preset time period, and the second user information is user information corresponding to the second meter information, and the second meter information and second user information are used as second data sources; At least based on specific information in the first user information and the second user information, cluster analysis is performed on the first data source and the second data source to generate an expert knowledge base.

8. The meter selection determination and intelligent recommendation method based on statistical algorithm according to claim 7 is characterized in that: At least according to the instantaneous flow value prediction upper limit and / or the instantaneous flow value prediction lower limit of the user's initial meter, recommending a target meter that can match the user from the target meters in the expert knowledge base, including: Expanding the upper limit of the instantaneous flow value prediction of the user's initial meter, and expanding the lower limit of the instantaneous flow value prediction of the user's initial meter to obtain an expanded flow range; At least according to the expanded flow interval range and the specific information of the user, a target meter that can match the user is recommended from the target meters in the expert knowledge base.

9. The meter selection determination and intelligent recommendation method based on statistical algorithm according to any one of claims 1 to 3, characterized in that: For multiple records of the first historical data set, determining invalid records in the multiple records based at least on the first cumulative flow value in the multiple records, deleting the invalid records in the multiple records, and obtaining a second historical data set, including: For multiple records of the first historical data set, determine a second cumulative flow value in each record; wherein the second cumulative flow value is obtained by performing a differential calculation on two first cumulative flow values ​​in two adjacent records; Records in which the second cumulative flow value is less than the set flow are recorded as invalid records, and the invalid records in the plurality of records are deleted to obtain a second historical data set.

10. The meter selection determination and intelligent recommendation method based on statistical algorithm according to claim 9, characterized in that: Each record in the first historical data set also includes environmental parameters, which include temperature and pressure. In the method, in addition to recording the record in which the second cumulative flow value is less than the set flow as an invalid record, the record in which the temperature exceeds the temperature threshold and / or the pressure exceeds the pressure threshold is also recorded as an invalid record.

11. A computer device, characterized in that: include: Memory for storing computer programs; A processor is used to execute the computer program, and when the computer program is executed by the processor, the steps of the meter selection determination and intelligent recommendation method based on statistical algorithm as described in any one of claims 1 to 10 are implemented.

Images