A method for evaluating loom operation efficiency

Abstract

The invention relates to the technical field of prediction and evaluation of loom operation efficiency, and in particular to a method for evaluating loom operation efficiency, comprising: collecting various data affecting the loom operation efficiency for multiple consecutive days, standardizing and processing the data into various indicators, and for each indicator: evaluating the relative discreteness of each indicator value, determining the abnormality of the indicator value according to the relative size between the relative discreteness of each indicator value and the average of the relative discreteness of multiple adjacent indicator values, performing smoothing operations on all indicator values ​​of the indicator and the abnormality of all indicator values, respectively obtaining the predicted indicator value and the predicted abnormality of the indicator, and adjusting the predicted indicator value using the predicted abnormality, taking the average of the adjusted predicted indicator values ​​of all indicators as the operation efficiency index of the loom, and evaluating the operation efficiency of the loom according to the operation efficiency index, thereby improving the accuracy and reliability of the prediction result and the accuracy of the operation efficiency evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of prediction and evaluation of loom operation efficiency, and in particular to a method for evaluating loom operation efficiency. Background Art

[0002] In the modern textile industry, looms are the core equipment in the production process, and their operating efficiency directly affects the production capacity of the enterprise. In the textile process, after collecting the actual operating data of the looms over a period of time, these data are used to analyze the possible operating status of the looms and evaluate the operating efficiency of the looms. The significance lies in being able to discover problems and make timely adjustments. If a loom is evaluated to be inefficient, timely maintenance or replacement of parts can be carried out to avoid production interruptions, optimize production plans, and ensure the smooth implementation of production plans.

[0003] Exponential smoothing is a prediction method based on time series, which predicts future data through historical data that has been collected to perform various predictions and evaluations. For example, the Chinese patent application document with the existing application publication number CN107122880A provides a method for predicting the trend of power equipment alarm information based on exponential smoothing, including: counting the number of historical alarm information of each substation, taking days as the unit, counting the number of alarm information of each substation as the observation value, determining the initial value of the exponential smoothing method according to the number of days of historical information observation, selecting the smoothing coefficient according to the trend of historical data, establishing an exponential smoothing prediction model, and predicting the possible value of the number of alarm information of the substation in the future, so as to improve the early warning capability of equipment operation trend.

[0004] When using the above-mentioned exponential smoothing method to predict and evaluate the operating efficiency of the loom based on the collected indicators that affect the loom, sudden equipment failures or improper operations will produce abnormal indicator values ​​that are significantly different from those during normal operation of the loom. These abnormal indicator values ​​cannot reflect the normal operating conditions of the loom. Abnormal indicator values ​​will cause the results predicted by the exponential smoothing method to deviate from the actual trend, reducing the accuracy of the exponential smoothing operation, and further affecting the accuracy of the evaluation of the loom's working efficiency. Summary of the invention

[0005] In order to solve the problem that the abnormal index value reduces the accuracy of the exponential smoothing operation, resulting in the inability to accurately evaluate the working efficiency of the loom, the present invention proposes a loom operation efficiency evaluation method, comprising:

[0006] Collect various data affecting the operating efficiency of the loom for multiple consecutive days, and perform standardization on the various data to obtain multiple index values ​​of corresponding indicators;

[0007] For each indicator, the relative dispersion of the indicator value is determined according to the size of the serial number of each indicator value and the overall deviation level of the indicator value and all indicator values. The relative dispersion is positively correlated with the size of the serial number and the overall deviation level. The serial number is determined based on the collection time of all indicator values. The abnormality of the indicator value is determined according to the relative size between the relative dispersion of each indicator value and the mean of the relative dispersion of its multiple adjacent indicator values;

[0008] Exponential smoothing operation is performed on all indicator values ​​of the indicator and the abnormality degree of all indicator values ​​to obtain the predicted indicator value of the indicator and the predicted abnormality degree of the indicator respectively, and the predicted indicator value of the indicator is adjusted using the predicted abnormality degree of the indicator; the predicted indicator value and the predicted abnormality degree are used to reflect the overall level of all indicator values ​​in a continuous number of days and the overall level of the abnormality degree of all indicator values ​​in a continuous number of days respectively;

[0009] The average of the adjusted predicted index values ​​of all the indexes is used as the operation efficiency index of the loom in a plurality of consecutive days, and the operation efficiency of the loom is evaluated according to the size of the operation efficiency index.

[0010] The above technical solution can obtain enough samples by collecting data for multiple consecutive days, which can more comprehensively and accurately reflect the operating status of the loom. The standardization process ensures the consistency of the dimensions between different indicators, which is convenient for subsequent data analysis and comparison. By calculating the relative discreteness of each indicator value, it is possible to identify which indicator values ​​deviate from the normal range. Further, by comparing the relative discreteness of a single indicator value and its adjacent indicator values ​​relative to all indicator values, the abnormality of each indicator value can be obtained. The greater the abnormality, the more likely the indicator value is abnormal. By performing double exponential smoothing operations on all indicator values ​​of each indicator that has been collected and the abnormality of all indicator values, the stability of the prediction results is improved, the influence of accidental factors is reduced, and the predicted indicator value and predicted abnormality of each indicator are obtained. The predicted indicator value is adjusted according to the predicted abnormality to make it closer to the actual change trend of the predicted indicator, reducing the influence of abnormal indicator values ​​on the prediction results and improving the accuracy of the exponential smoothing operation. Finally, the predicted indicator values ​​of all indicators are quantified into a comprehensive operating efficiency index. The operating efficiency index comprehensively reflects the overall operating conditions of the loom in multiple consecutive days, and accurately quantifies the operating efficiency of the loom, which is convenient for timely management decisions.

[0011] Preferably, the method of adjusting the predicted index value of the index using the predicted abnormality degree of the index is:

[0012] If the predicted abnormality of the indicator is greater than the preset abnormality threshold, the adjusted predicted indicator value of the indicator satisfies the following relationship:

[0013]

[0014] In the formula, For the The adjusted predicted indicator value of each indicator, For the The predicted abnormality of each indicator, is the preset abnormality threshold, For the The predicted indicator value of the indicator, For the The mean of all indicator values ​​of an indicator;

[0015] If the predicted abnormality level of the indicator is not greater than the preset abnormality level threshold, the adjusted predicted indicator value of the indicator is equal to the predicted indicator value of the indicator.

[0016] The above technical solution effectively reduces the impact of abnormal indicator values ​​on the prediction results by adjusting the future indicator values ​​to be closer to the historical average, thereby improving the accuracy and reliability of the future indicator values. Specifically, when the degree of abnormality of the future indicator values ​​exceeds the preset threshold, it will be corrected to varying degrees according to the degree of abnormality to make it more in line with the normal operating trend. This method helps to improve the accuracy of the evaluation of the loom's operating efficiency and support more reasonable production plans and maintenance decisions.

[0017] Preferably, the various data affecting the operating efficiency of the loom include: daily output, operating time ratio, downtime, and scrap rate; wherein the operating time ratio is the ratio of the actual operating time of the loom to the planned operating time of the loom, and the scrap rate is the ratio of the amount of scrap generated to the total output.

[0018] Preferably, the method for evaluating the operating efficiency of the loom according to the magnitude of the operating efficiency index is:

[0019] Calculate the mean and standard deviation of the operating efficiency index of all looms in multiple consecutive days, and use the difference between the mean and the standard deviation as the operating efficiency threshold;

[0020] If the operating efficiency index of a loom is lower than the operating efficiency threshold, the operating efficiency of the loom is low and needs to be adjusted; if the operating efficiency index of a loom is not lower than the operating efficiency threshold, the operating efficiency of the loom is normal and does not need to be adjusted.

[0021] The above technical solution provides an objective standard to evaluate the operating efficiency of each loom, which helps to identify inefficient looms in a timely manner and take corresponding measures, thereby improving overall production efficiency and equipment management level.

[0022] Preferably, the abnormality degree of the indicator value is the ratio of the relative dispersion degree of the indicator value to the mean of the relative dispersion degrees of its multiple adjacent indicator values.

[0023] The above technical solution can measure the degree of deviation of a single indicator value from the overall data distribution by calculating the relative dispersion of the single indicator value, and compare the relative dispersion of the single indicator value with the mean relative dispersion of adjacent indicator values ​​to further determine the relative deviation of the indicator value in the local time series. By comparing with adjacent indicator values, it can reduce misjudgment caused by random fluctuations or short-term changes and improve the accuracy of anomaly detection.

[0024] Preferably, the method of performing standardization processing on the data to obtain multiple indicator values ​​of the corresponding indicators is:

[0025] Normalize the data for multiple consecutive days respectively so that the data are Within the scope, the hierarchical analysis method is used to determine the weight of each data, and the normalized value of each data is weighted with its weight as each indicator.

[0026] Preferably, the relative dispersion of each indicator value satisfies the following relationship:

[0027]

[0028] In the formula, For the The first indicator The relative dispersion of the index values, is the serial number of each indicator value of this indicator, is the total number of all indicator values ​​of this indicator, For the The first indicator The overall deviation level of the indicator value from all indicator values, and , For the The first indicator The index value, For the The mean of all indicator values ​​of an indicator. For the The standard deviation of all indicator values ​​of an indicator, The smoothing exponent of the preset exponential smoothing method.

[0029] The above technical solution combines the time factor weight of each indicator value and the degree of deviation of each indicator value from all indicator values ​​as a whole to quantify the relative discreteness of each indicator value. The greater the degree of deviation, the more likely the indicator value is to be abnormal. Considering that all indicator values ​​contained in each indicator are obtained based on the order of collection time, the position of each indicator value of the indicator in the time series is analyzed. The time factor weight of each indicator value is introduced, so that the recent indicator value has a higher weight and the early indicator value has a lower weight, because the recent indicator value tends to better reflect the current operating status and trend of the loom. By giving higher weights to recent data, the latest changes and abnormal situations can be captured more timely, which can improve the accuracy of abnormal analysis.

[0030] Preferably, the multiple adjacent index values ​​of each index value are selected from the index values ​​that are closest to the index value in terms of collection time. index values, as multiple adjacent index values ​​of the index value, is the default value.

[0031] Preferably, the frequency of collection is once a day.

[0032] The present invention has the following effects:

[0033] The present invention collects various indicators for multiple consecutive days to comprehensively and accurately reflect the operating status of the loom, calculates the relative discreteness and abnormality of each indicator value, effectively identifies the indicator value that deviates from the normal range, and obtains the predicted indicator value and predicted abnormality of each indicator by performing double exponential smoothing operations on the indicator value and its abnormality. The predicted indicator value is corrected using the predicted abnormality, thereby improving the stability and accuracy of the prediction result, reducing the impact of abnormal data, and improving the accuracy of the evaluation result of the operating efficiency of the loom. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0035] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0037] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0038] Reference Figure 1 The loom operation efficiency evaluation method provided by the present invention comprises steps S1 to S5:

[0039] S1: Collect various data that affect the operating efficiency of the loom, standardize the data, and obtain the corresponding index values ​​of various indicators.

[0040] During the operation of the loom, the various data that affect its operating efficiency include:

[0041] Daily output: record the total quantity produced by each loom every day, the unit can be pieces or meters, etc.;

[0042] Running time ratio: the ratio of the actual running time of the loom to the planned running time of the loom. For example, if the planned working time is 8 hours and the actual working time is 7 hours, the ratio is 0.875;

[0043] Downtime: record the downtime caused by machine failure or other reasons, in minutes or hours;

[0044] Scrap rate: Calculate the ratio of scrap produced in the daily production process to the total output. For example, if the total output is 1,000 pieces and there are 50 scrap pieces, the scrap rate is 5%.

[0045] Therefore, this step needs to collect these four data during the operation of the loom and record them continuously. days, and record once a day, then 4 data items are collected every day, and each data item has only one value, among which, The value is 30, which is an empirical value and can be adjusted according to actual needs.

[0046] Next, the data collected over multiple consecutive days are standardized, specifically by normalizing and weighting each data item:

[0047] Data normalization: Since the data have different orders of magnitude, for example, the output may be hundreds of pieces, while the scrap rate may be a few percent. Directly using these data for evaluation will cause the impact of certain data to be magnified or reduced. Therefore, first use normalization operations to unify all data to the same order of magnitude, usually between 0 and 1. This can eliminate the impact of different dimensions and facilitate subsequent analysis and processing.

[0048] Weight analysis: Even if the data has been normalized, different data items still have different impacts on the operating efficiency of the loom. For example, downtime may have a greater impact on operating efficiency than scrap rate. Therefore, the analytic hierarchy process is used to determine the weight of each data item. The analytic hierarchy process is a multi-criteria decision-making method that can systematically evaluate the importance of each factor and assign weights to each factor to reflect its impact on operating efficiency. The greater the weight of a data item, the greater the impact of the data item on operating efficiency.

[0049] Next, multiple index values ​​of various indicators corresponding to various data are obtained according to the normalization results of various data and the weight analysis results. The specific operation is: multiply the normalized values ​​of various data by the corresponding weights, and use the multiplied values ​​as the index values ​​of various indicators. Specifically, the normalized value of output is multiplied by the weight of output to obtain the index value of output index, the normalized value of operation time ratio is multiplied by the weight of operation time ratio to obtain the index value of operation time ratio index, the normalized value of downtime is multiplied by the weight of downtime to obtain the index value of downtime index, and the normalized value of scrap rate is multiplied by the weight of scrap rate to obtain the index value of scrap rate index.

[0050] According to this method, the index value of each indicator for each day (the index value of the output index, the index value of the operating time ratio index, the index value of the downtime index, the index value of the scrap rate index) is calculated. Since the data are collected for a total of 30 days, each indicator includes 30 index values.

[0051] Through the standardization process in this step, the index values ​​of various indicators obtained can more accurately evaluate the operating efficiency of the loom.

[0052] S2: For each indicator, evaluate the relative dispersion of each indicator value, and determine the abnormality of the indicator value based on the relative dispersion.

[0053] During the operation of the loom, due to machine failure or operating errors, extreme abnormal values ​​may occur, such as the output on a certain day is suddenly very low or very high. If these data containing abnormal values ​​are directly used to evaluate the operating efficiency, the evaluation results will be inaccurate and even mislead the decision. Therefore, in order to reduce the impact of abnormal data, this step obtains the abnormality of each indicator value based on the relative discreteness of each indicator value to reflect the possibility of abnormality of the indicator value.

[0054] Among them, the relative discreteness of each indicator value satisfies the following relationship:

[0055]

[0056] In the formula, For the The first indicator The relative dispersion of the index values, is the serial number of each indicator value of this indicator, is the total number of all indicator values ​​of this indicator, For the The first indicator The overall deviation level of the indicator value from all indicator values, and , For the The first indicator The index value, For the The mean of all indicator values ​​of an indicator. For the The standard deviation of all indicator values ​​of an indicator, The smoothing index of the preset exponential smoothing method. This step presets the smoothing index , this value is an empirical value and can be set by yourself.

[0057] In this formula, Partly by introducing the time factor weight of each indicator value, when When smaller, The value is close to ,when near hour, The value is close to , The greater the time factor weight of an indicator value, the greater the impact of the indicator value on the evaluation process of the loom's operating efficiency, because all indicator values ​​of each indicator are collected daily, which is a time series. The earlier the collection time of a certain indicator value, the smaller its impact on the evaluation process of the loom's operating efficiency, and the smaller the time factor weight of the indicator value. Conversely, if a certain indicator value is collected recently, the greater its impact on the evaluation process of the loom's operating efficiency, the greater the time factor weight of the indicator value. Through this operation, the recent indicator value has a higher time factor weight, and the early indicator value has a lower time factor weight, because the recent indicator value often better reflects the current operating status and trend of the loom, and can capture the latest changes and abnormal conditions more timely.

[0058] In this formula, Part means first calculate the The difference between the first index value and the average index value is calculated, and then the ratio of the difference to the standard deviation of the index value is calculated. The larger the ratio is, the better the The greater the gap between the individual index values ​​and all the index values ​​as a whole, the The larger the deviation of an indicator value, the more likely it is an abnormal indicator value.

[0059] In summary, if the collection time of an indicator value is closer to the current evaluation time (the position of the indicator value in the time series is later, that is, The larger the value of ), and the larger the overall gap between this index value and all index values, the greater the influence of this index value on the evaluation result of the loom operation efficiency. That is, the greater the relative discreteness of the index value, the easier it is to affect the accuracy of the evaluation result of the loom operation efficiency.

[0060] Among them, the method for obtaining the abnormality degree of each indicator value is:

[0061] First select the value closest to the index value in terms of collection time. index values, as multiple adjacent index values ​​of the index value, including the index value before / 2 indicator values ​​and the value after the indicator value / 2 indicator values, is a preset value, which is set to 20 in this step. This value is an empirical value and can be adjusted. For example, for the index value collected on the 11th day, the index values ​​collected from the 1st to the 10th day and the index values ​​collected from the 12th to the 21st day are used as the adjacent index values ​​of the index value collected on the 11th day.

[0062] Then calculate this The average of the relative dispersion of adjacent index values ​​is used to determine the abnormality of the index value according to the relative size between the relative dispersion of the index value and the average of the dispersion. The specific operation is as follows:

[0063] The ratio of the relative dispersion of the index value to the mean of the dispersion is taken as the abnormality of the index value, which can be expressed as follows:

[0064]

[0065] In the formula, Indicates Among the indicators The abnormality of the indicator value, Indicates The first indicator The relative dispersion of the index values, is a preset number, indicating the total number of adjacent indicator values, Indicates Among the indicators The index value of The relative dispersion of adjacent indicator values.

[0066] The denominator in the formula is Among the indicators The average of the relative dispersion of multiple adjacent index values ​​of an index value, and the numerator represents the Among the indicators The relative dispersion of the index values. This ratio reflects the relative size of the two. The larger the ratio, the The higher the relative dispersion level of an indicator value, the more likely it is an abnormal indicator value.

[0067] S3: The predicted index value of each indicator and the predicted abnormality degree of each indicator are obtained through exponential smoothing method.

[0068] If all the index values ​​of each indicator are directly used to evaluate the operating efficiency of the loom, it may be affected by noise and outliers, resulting in inaccurate evaluation results. Therefore, this step performs exponential smoothing on all the index values ​​of each indicator to obtain the predicted index value of the indicator. The exponential smoothing method reduces the impact of noise and outliers through smoothing, thereby obtaining a more stable predicted index value. The predicted index value is used to reflect the overall level of all the index values ​​over multiple consecutive days. Next, the abnormality of all the index values ​​of each indicator is also subjected to exponential smoothing to obtain the predicted abnormality of the indicator. The predicted abnormality reflects the overall level of the abnormality of all the index values ​​over multiple consecutive days.

[0069] S4: Use the predicted abnormality of each indicator to adjust the predicted indicator value of the indicator.

[0070] This step is equivalent to using the predicted abnormality of each indicator as a correction factor for the predicted indicator value of the indicator, using the correction factor to adjust the predicted indicator value of the indicator, and presetting an abnormality threshold for the indicator. , the implementers can set it according to the actual implementation situation In this embodiment, the value of .

[0071] If the predicted abnormality of the indicator is greater than , indicating that the predicted index value of this indicator is likely to be abnormal data, indicating that the abnormal data existing in all the index values ​​of this indicator has a greater impact on the prediction results of the exponential smoothing method. Therefore, it is necessary to use the predicted abnormality degree of this indicator as a correction factor to adjust the predicted index value of this indicator to obtain a more accurate predicted index value. Specifically, the predicted index value of this indicator is adjusted according to the following formula:

[0072]

[0073] In the formula, For the The adjusted predicted indicator value of each indicator, For the The predicted abnormality of each indicator, is the preset abnormality threshold, For the The predicted indicator value of each indicator (before adjustment), For the The mean of all indicator values ​​for an indicator.

[0074] If the predicted abnormality of the indicator is not greater than , indicating that the predicted indicator value of this indicator is less likely to be abnormal data, and the predicted indicator value of this indicator is directly used as the adjusted predicted indicator value.

[0075] S5: Evaluate the operating efficiency of the loom based on the adjusted predicted index values ​​of all item indicators.

[0076] Specifically, the mean of the adjusted predicted index values ​​of all indicators is used as the loom in the continuous The operation efficiency index within the day, that is, obtaining the predicted index value of the output index, the predicted index value of the operation time ratio index, the predicted index value of the downtime index, and the predicted index value of the scrap rate index, and taking the average value of these four predicted index values ​​as the operation efficiency index of the loom.

[0077] Next, the operating efficiency of the loom is evaluated according to the size of the operating efficiency index:

[0078] Calculate the mean and standard deviation of the operating efficiency index of all looms in multiple consecutive days, and use the difference between the mean and the standard deviation as the operating efficiency threshold; if the operating efficiency index of a loom is lower than the operating efficiency threshold, the operating efficiency of the loom is low and needs to be adjusted; if the operating efficiency index of a loom is not lower than the operating efficiency threshold, the operating efficiency of the loom is normal and does not need to be adjusted.

[0079] Among them, the value obtained by subtracting the standard deviation of the data from the mean of the data is of great significance in statistics, especially when evaluating data distribution and setting thresholds. Specifically, this value serves as a reference point to identify lower values ​​in the data. Therefore, this step uses the method of subtracting the standard deviation from the mean to calculate the operating efficiency threshold. Looms with operating efficiency lower than the operating efficiency threshold are considered to be inefficient and need to be adjusted. Looms with operating efficiency not lower than the operating efficiency threshold are considered to be normal and do not need to be adjusted. This helps to promptly discover and deal with inefficient equipment and improve overall production efficiency. Through continuous regular evaluation and adjustment, the operating status of the loom can be continuously monitored and optimized to ensure that the equipment is always in an efficient operating state.

[0080] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0081] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for evaluating the operating efficiency of a loom, characterized in that: include: Collect various data affecting the operating efficiency of the loom for multiple consecutive days, and perform standardization on the various data to obtain multiple index values ​​of corresponding indicators; For each indicator, the relative dispersion of the indicator value is determined according to the size of the serial number of each indicator value and the overall deviation level of the indicator value and all indicator values. The relative dispersion is positively correlated with the size of the serial number and the overall deviation level. The serial number is determined based on the collection time of all indicator values. The abnormality of the indicator value is determined according to the relative size between the relative dispersion of each indicator value and the mean of the relative dispersion of its multiple adjacent indicator values; Exponential smoothing operation is performed on all indicator values ​​of the indicator and the abnormality degree of all indicator values ​​to obtain the predicted indicator value of the indicator and the predicted abnormality degree of the indicator respectively, and the predicted indicator value of the indicator is adjusted using the predicted abnormality degree of the indicator; the predicted indicator value and the predicted abnormality degree are used to reflect the overall level of all indicator values ​​in a continuous number of days and the overall level of the abnormality degree of all indicator values ​​in a continuous number of days respectively; The average of the adjusted predicted index values ​​of all the indexes is used as the operation efficiency index of the loom in a plurality of consecutive days, and the operation efficiency of the loom is evaluated according to the size of the operation efficiency index.

2. The loom operation efficiency evaluation method according to claim 1, characterized in that: The method of adjusting the predicted index value of the index using the predicted abnormality of the index is as follows: If the predicted abnormality of the indicator is greater than the preset abnormality threshold, the adjusted predicted indicator value of the indicator satisfies the following relationship: ； In the formula, For the The adjusted predicted indicator value of each indicator, For the The predicted abnormality of each indicator, is the preset abnormality threshold, For the The predicted indicator value of the indicator, For the The mean of all indicator values ​​of an indicator; If the predicted abnormality level of the indicator is not greater than the preset abnormality level threshold, the adjusted predicted indicator value of the indicator is equal to the predicted indicator value of the indicator.

3. The loom operation efficiency evaluation method according to claim 1, characterized in that: The various data affecting the operating efficiency of the loom include: daily output, operating time ratio, downtime, and scrap rate; wherein the operating time ratio is the ratio of the actual operating time of the loom to the planned operating time of the loom, and the scrap rate is the ratio of the amount of scrap generated to the total output.

4. The loom operation efficiency evaluation method according to claim 1, characterized in that: The method for evaluating the operating efficiency of the loom according to the size of the operating efficiency index is: Calculate the mean and standard deviation of the operating efficiency index of all looms in multiple consecutive days, and use the difference between the mean and the standard deviation as the operating efficiency threshold; If the operating efficiency index of a loom is lower than the operating efficiency threshold, the operating efficiency of the loom is low and needs to be adjusted; If the operating efficiency index of a loom is not lower than the operating efficiency threshold, the operating efficiency of the loom is normal and does not need to be adjusted.

5. The loom operation efficiency evaluation method according to claim 2, characterized in that: The abnormality of an indicator value is the ratio of the relative dispersion of the indicator value to the mean of the relative dispersion of its multiple adjacent indicator values.

6. The loom operation efficiency evaluation method according to claim 3, characterized in that: The method of performing standardization processing on the data to obtain multiple index values ​​of the corresponding indicators is: Normalize the data of consecutive days separately so that each data is in Within the scope, the hierarchical analysis method is used to determine the weight of each data, and the normalized value of each data is weighted with its weight as the indicator value of each indicator corresponding to each data.

7. The method for evaluating loom operation efficiency according to claim 5, characterized in that: The relative dispersion of each indicator value satisfies the following relationship: ； In the formula, For the The first indicator The relative dispersion of the index values, is the serial number of each indicator value of this indicator, is the total number of all indicator values ​​of this indicator, For the The first indicator The overall deviation level of the indicator value from all indicator values, and , For the The first indicator The index value, For the The mean of all indicator values ​​of an indicator. For the The standard deviation of all indicator values ​​of an indicator, The smoothing exponent of the preset exponential smoothing method.

8. The method for evaluating loom operation efficiency according to claim 5, characterized in that: The multiple adjacent index values ​​of each index value are selected from the index value that is closest to the index value in terms of collection time. index values, as multiple adjacent index values ​​of the index value, is the default value.

9. The method for evaluating loom operation efficiency according to claim 8, characterized in that: The frequency of collection is once a day.

Images