CN-121996899-A - Consumption limit calculation method for equipment

CN121996899ACN 121996899 ACN121996899 ACN 121996899ACN-121996899-A

Abstract

The invention relates to the technical field of data evaluation, in particular to a calculation method for consumption limit of equipment, which comprises the steps of obtaining historical consumption data and a historical consumption predicted value of equipment with any consumption limit to be evaluated, determining the number of times of an exponential smoothing method according to the change condition of the historical consumption data and calculating the consumption of the equipment by using the exponential smoothing method, calculating historical fluctuation data based on the difference value of the historical consumption data and the historical consumption predicted value, estimating the deviation amount of the consumption of the equipment of the current year based on an uncertainty theory, and adjusting the consumption of the equipment determined according to the exponential smoothing method according to the deviation amount of the consumption of the equipment as the consumption of the equipment and evaluating a deviation credible interval based on the historical fluctuation data. The invention determines the consumption and deviation credible interval of equipment based on an exponential smoothing method and an uncertainty theory, and simultaneously solves two problems of consumption and possible fluctuation evaluation.

Inventors

WEN GUOYI
DENG CHAO
YANG MING
CHEN WANSHE
CUI HONGXIN
TIAN ZEHUA
YANG XUDONG

Assignees

中国人民武装警察部队研究院

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (7)

1. A method for calculating consumption limit of equipment is characterized by comprising the following steps, Acquiring the annual consumption data and the annual consumption predicted value of equipment of any consumption to be evaluated; Determining the number of times of an exponential smoothing method according to the change condition of the consumption data in the past year, and calculating the consumption of equipment by using the exponential smoothing method based on the determined number of times of the exponential smoothing method; calculating to form annual fluctuation data by using the difference value between the annual consumption data and the annual consumption predicted value, estimating the deviation amount of the consumption amount of equipment of the present year based on an uncertainty theory, and adjusting the consumption amount according to the deviation amount; constructing a deviation credible interval based on the annual fluctuation data by taking consumption as a center; wherein the consumption limit is a set comprising consumption and deviation confidence intervals.
2. The equipment consumption amount calculation method according to claim 1, wherein the equipment consumption amount is calculated by an exponential smoothing method based on the past year consumption data.
3. The equipment consumption calculating method according to claim 2, wherein the number of times of the exponential smoothing method is determined in accordance with fluctuation conditions of the past-year consumption data; if the fluctuation of the annual consumption data fluctuates in a preset interval, namely the adjacent value deviation of the annual consumption data is within the maximum value of the absolute value of the annual fluctuation data, judging that the fluctuation is gentle, and adopting a primary exponential smoothing method; if the fluctuation of the historical consumption data is not in the preset interval, judging the number of exponential smoothing methods according to the change trend of the historical consumption data; If the consumption data in the past year is in a unidirectional increasing trend or a unidirectional decreasing trend, adopting a secondary exponential smoothing method; If the consumption data in the past year is in a curve change trend, adopting a three-time exponential smoothing method.
4. The equipment consumption amount calculation method according to claim 2, wherein the process of determining the calculation initial value of the exponential smoothing method includes: If the cumulative data years after the equipment is shaped exceeds ten years, selecting an average value of the annual consumption data of the equipment before ten years as a calculation initial value of an exponential smoothing method, and performing iterative calculation on the annual consumption data of the equipment in ten years based on the exponential smoothing method; If the cumulative data age of the equipment is less than or equal to ten years, the calendar average value of similar equipment can be used as a calculation initial value, and the calendar average value is weighted and adjusted according to the characteristic difference of the selected equipment; If the equipment of similar equipment is not available, an expert scoring method is adopted to estimate the initial value.
5. The equipment consumption limit calculation method according to claim 1, wherein the process of estimating the deviation amount of the annual equipment consumption amount based on the uncertainty theory includes; the deviation amount calculation of the consumption amount is performed based on the uncertainty theory according to the chronology fluctuation data, which is the difference between the chronology consumption data and the corresponding chronology consumption predicted value.
6. The equipment consumption limit calculation method according to claim 5, wherein the process of calculating the deviation amount of the annual equipment consumption amount from the annual fluctuation data includes: ① Extracting the maximum value, the minimum value and the mode of the annual fluctuation data; ② Constructing a triangular fuzzy variable based on the maximum value and the maximum value of the deviation between the minimum value and the mode of the annual fluctuation data; ③ The method comprises the steps of performing discretization processing on triangular fuzzy variables based on a fuzzy discretization method, dividing a supporting interval into a plurality of subintervals, dividing subintervals by using age fluctuation data as deviation values in deviation amount prediction of equipment consumption, and determining credibility measure of each deviation value according to occurrence probability of the deviation value; ④ Taking the maximum value and the probability maximum value of the discrete fuzzy variable of each support interval to determine the discrete fuzzy variable and the credibility measure thereof, and determining the equivalent value of the discrete fuzzy variable; ⑤ The equivalent value of the discrete blur variable is used as the deviation amount of the consumption amount.
7. The consumption limit calculation method of equipment according to claim 1, wherein the process of determining the deviation confidence interval satisfying the consumption limit with the confidence measure as the standard value is: ① Calculating the root mean square of the annual fluctuation data; ② Rounding up the root mean square of the annual fluctuation data; ③ Constructing a floating interval according to the upper rounding; ④ Counting the proportion of the annual fluctuation data which do not belong to the floating interval; ⑤ If the proportion of the calendar fluctuation data which does not belong to the floating interval does not meet the set credibility measure standard numerical range, expanding an interval by one unit until the proportion of the calendar fluctuation data which does not belong to the floating interval is within the preset credibility measure standard numerical range, and determining the floating interval as a deviation credibility interval.

Description

Consumption limit calculation method for equipment Technical Field The invention relates to the technical field of data evaluation, in particular to a consumption limit calculation method for equipment. Background At present, the consumption limit calculation of equipment is mainly divided into three types, namely a time sequence prediction method, prediction based on machine learning, and combined prediction, namely, two or more methods are adopted for the same problem. These prediction methods are chosen based primarily on existing data, as they are all predicted using existing historical statistics or factor variable data. The time series prediction method mainly predicts future conditions according to the past time series data, and comprises a moving average method, an exponential smoothing method and the like. The moving average method is a method for predicting the demand of equipment in a future period or several periods by using a latest set of actual data, and further determining the ordered number, etc., and is suitable for on-demand prediction, and is very useful in that the moving average method can effectively eliminate random fluctuations in prediction when the demand of products is neither rapidly increased nor rapidly decreased and seasonal factors are not present, and has the disadvantages of being unsuitable for prediction of the trend of fluctuation of equipment, and also requiring reservation of data of the predicted past periods of each equipment, requiring a large amount of storage resources, and being troublesome in calculation. Another commonly used prediction method is an exponential smoothing method, which does not require so much data, and is simple and convenient to calculate, and more suitable for trend prediction. The prediction based on machine learning is a method for establishing a consumption prediction model by researching the correlation between variables through a modern machine learning algorithm and theory, carrying out mathematical statistics analysis and training fitting on sample data, and mainly comprises linear regression, boottrap, a neural network, a support vector machine, a gray prediction method and the like. In summary, based on machine learning prediction, it is necessary to provide a large amount of data under the influence of factors affecting equipment consumption of as full as possible, for example, a large amount of data under the long-term influence of factors such as the number of tasks, the use environment, the use level, the maintenance level, etc., which is difficult for many units at present, to require quantitative statistics and provision. In the exponential smoothing method, as a time sequence prediction method, the influence caused by random factors can be eliminated through the layer-by-layer smoothing calculation of historical statistical sequence data of a prediction target, and the basic change trend of the prediction target is found out and predicted in the future according to the basic change trend, so that the exponential smoothing method is a suitable method for calculating the consumption limit of equipment. At the same time, however, there is uncertainty as to what has not yet occurred. The prediction of equipment consumption is affected by a plurality of uncertain factors such as the number of future tasks, the use environment, the use level, the maintenance level and the like, uncertainty exists, and the accurate consumption must not be known until the occurrence of things. For example, if a certain unit loss tire number limit is given only by an exponential smoothing method in the present year, there is a problem that the consumption limit is small and cannot be purchased, the consumption limit is large, and the number of other equipment is limited when the total amount is limited. Disclosure of Invention Therefore, the invention provides a method for calculating the consumption limit of equipment, which is used for solving the problem that the consumption limit of equipment is evaluated unreasonably in the prior art, so that the purchasing quantity of the equipment is limited unreasonably. In order to achieve the above object, the present invention provides a method for calculating consumption limit of equipment, comprising: Acquiring the annual consumption data and the annual consumption predicted value of equipment of any consumption to be evaluated; Determining the number of times of an exponential smoothing method according to the change condition of the consumption data in the past year, and calculating the consumption of equipment by using the exponential smoothing method based on the determined number of times of the exponential smoothing method; Calculating to form historical fluctuation data by using the difference value between the historical consumption data and the historical consumption predicted value, estimating the deviation amount of the consumption amount of equipment of the current year based on an uncertainty