Method for calculating consumption limit of equipment

Abstract

The invention relates to the technical field of data evaluation, in particular to an equipment consumption limit calculation method, which comprises the following steps of: acquiring over-the-year consumption data and over-the-year consumption estimated value of any equipment of which the consumption limit is to be evaluated; determining the number of times of an exponential smoothing method according to the change condition of the consumption data over the years, and calculating the consumption of the equipment by using the exponential smoothing method; calculating fluctuation data over the years based on a difference value between the consumption data over the years and a consumption estimated value over the years, and estimating a deviation value of the consumption of the equipment in the current year based on an uncertainty theory; according to the deviation value of the consumption of the equipment, the consumption of the equipment determined according to the exponential smoothing method is adjusted as the consumption of the equipment, and a deviation credible interval is evaluated based on fluctuation data over the years. According to the method, the consumption and the deviation credible interval of the equipment are determined based on the exponential smoothing method and the uncertainty theory, and meanwhile the two problems of consumption and possible fluctuation evaluation are solved.

Description

Technical Field

[0001] This invention relates to the field of data evaluation technology, and in particular to a method for calculating the consumption limit of equipment. Background Technology

[0002] Currently, the calculation of equipment consumption limits mainly falls into three categories: time series forecasting, machine learning-based forecasting, and combined forecasting, which uses two or more methods to predict the same problem. The selection of these forecasting methods is primarily based on existing data, as they all utilize historical statistical data or factor variable data for prediction. Time series forecasting mainly predicts future conditions based on past time series data, including moving averages and exponential smoothing. Moving averages use the most recent set of actual data to predict the demand for equipment in the next one or several periods, thereby determining the order quantity. This method is suitable for short-term forecasting. When product demand neither grows nor declines rapidly, and there are no seasonal factors, moving averages can effectively eliminate random fluctuations in the forecast, making them very useful. However, their disadvantages include unsuitability for forecasting equipment with fluctuating trends, the need to retain data from multiple periods for each type of equipment being predicted (requiring significant storage resources), and computational complexity. Another commonly used forecasting method is exponential smoothing, which requires less data, is simpler to calculate, and is more suitable for trend-based forecasting. Machine learning-based prediction is a method that uses modern machine learning algorithms and theories to study the correlation between variables, perform mathematical and statistical analysis and training on sample data, and establish a predictive model. It mainly includes linear regression, bootstrap, neural networks, support vector machines, and grey prediction methods.

[0003] In summary, machine learning-based predictions require a large amount of data on all factors influencing equipment consumption, such as the long-term impact of factors like mission volume, usage environment, usage level, and maintenance level. For many organizations, quantifying and providing this data is currently difficult. Exponential smoothing, as a time series forecasting method, can eliminate the influence of random factors by performing layer-by-layer smoothing calculations on the historical statistical sequence data of the target, identifying the basic trend of the target and using this to predict the future. This makes it a suitable method for calculating equipment consumption limits.

[0004] However, there is uncertainty regarding events that have not yet occurred. The prediction of equipment consumption is influenced by many uncertain factors, such as the number of future missions, the usage environment, the level of use, and the level of maintenance. Therefore, the accurate consumption amount cannot be known until the event occurs. For example, if a unit's tire consumption limit for this year is given solely based on exponential smoothing, then there is the problem that "if the consumption limit is too low, we can't buy more; if the consumption limit is too high, the quantity of other equipment will be limited given the total budget." Summary of the Invention

[0005] To address this issue, the present invention provides a method for calculating the consumption limit of equipment, thereby overcoming the problem that the existing technology's assessment of the consumption limit of equipment is unreasonable, resulting in unreasonable restrictions on the purchase quantity of equipment.

[0006] To achieve the above objectives, the present invention provides a method for calculating the consumption limit of equipment, comprising: Obtain historical consumption data and historical consumption estimates for any equipment category whose consumption needs to be evaluated. The number of exponential smoothing operations is determined based on the changes in consumption data over the years. Based on the determined number of exponential smoothing operations, the consumption of equipment is calculated using the exponential smoothing method. Historical fluctuation data is calculated by the difference between historical consumption data and historical consumption estimates. Based on uncertainty theory, the deviation of the current year's equipment consumption is estimated, and the consumption is adjusted according to the deviation of the equipment consumption. Based on historical fluctuation data, a reliable range of deviations is constructed with consumption as the central factor. The consumption limit is a set that includes the consumption amount and the confidence interval of the deviation.

[0007] Furthermore, the consumption of equipment was calculated using the exponential smoothing method based on historical consumption data.

[0008] Furthermore, the number of exponential smoothing operations is determined based on the fluctuations in consumption data over the years; If the fluctuations in the annual consumption data are within a preset range, that is, if the deviation between adjacent values ​​of the annual consumption data is within the maximum value of the absolute value of the annual fluctuation data, then the fluctuations are determined to be gentle, and the first exponential smoothing method is adopted. If the fluctuations in the annual consumption data are not within the preset range, the number of exponential smoothing operations will be determined based on the changing trend of the annual consumption data. If the historical consumption data shows a unidirectional increasing trend or a unidirectional decreasing trend, the quadratic exponential smoothing method should be used; If the historical consumption data shows a curvilinear trend, the triple exponential smoothing method should be used. In practical calculations, abnormal data caused by special reasons can be deleted. At the same time, the presence or absence of trend and seasonality can be used as a preliminary judgment in conjunction with the traditional exponential smoothing method. That is, if there is no trend and no seasonality, choose the single exponential smoothing method; if there is a trend but no seasonality, choose the double exponential smoothing method; and if there is a trend and seasonality, choose the triple exponential smoothing method.

[0009] The exponential smoothing order mentioned in this invention is also referred to as the "order" in some other literature.

[0010] Furthermore, the process of calculating the initial value using the exponential smoothing method includes: If the cumulative data period after the equipment type is finalized exceeds ten years, the average value of the annual consumption data of the equipment type ten years ago is selected as the initial value for the exponential smoothing method, and the annual consumption data of the equipment type in the past ten years is iteratively calculated based on the exponential smoothing method. If the cumulative data period of the equipment category is less than or equal to ten years, the historical average of similar equipment categories can be used as the initial value for calculation, and the historical average can be weighted and adjusted according to the differences in the characteristics of the selected equipment category. If there is no similar equipment, the initial value is estimated using the expert scoring method.

[0011] Furthermore, the process of predicting the deviation of equipment consumption for this year based on uncertainty theory includes: The deviation in consumption is calculated based on historical fluctuation data and uncertainty theory. The historical fluctuation data is the difference between the historical consumption data and the corresponding historical consumption estimate.

[0012] Furthermore, the process of calculating the deviation of equipment consumption for the current year based on historical fluctuation data includes: ① Extract the maximum, minimum, and mode of the historical fluctuation data; ② Based on the mode, determine the maximum value of the deviation between the maximum and minimum values ​​and the mode, and construct a triangular fuzzy variable; ③ Based on the fuzzy discretization method, the triangular fuzzy variables are discretized and the support interval is divided into several sub-intervals. In the prediction of the deviation of equipment consumption, the historical fluctuation data is used as the deviation value to divide the sub-intervals, and the reliability measure of each deviation value is determined according to the probability of the deviation value. ④ Take the maximum value and probability maximum value of the discrete fuzzy variable in each support interval to determine the discrete fuzzy variable and its credibility measure, and determine the equivalent value of the discrete fuzzy variable; ⑤ Use the equivalent value of discrete fuzzy variables as the deviation of consumption.

[0013] Furthermore, determine the reliable interval of the deviation of the consumption limit that meets the reliableness measure as the standard value; ① Calculate the root mean square of the historical fluctuation data; ②Round up the root mean square of the historical fluctuation data; ③ Construct a floating range based on the floor function; ④ Calculate the proportion of historical fluctuation data that does not fall within the fluctuation range; ⑤ If the proportion of historical fluctuation data that does not belong to the floating range does not meet the set reliability measurement standard value range, then expand the floating range by one unit until the proportion of historical fluctuation data that does not belong to the floating range is within the preset reliability measurement standard value range, then determine the floating range as the deviation reliability range.

[0014] Compared with existing technologies, the beneficial effects of this invention are that the consumption of equipment based on exponential smoothing and uncertainty theory can serve as a reference for the quantity of equipment allocated to units; and the deviation confidence interval determined based on historical fluctuation data and confidence measures can serve as a basis for procurement fluctuations. In this way, it does not affect the estimation of consumption and its value, and it solves the practical problem of insufficient fluctuation space after consumption is fixed, making it difficult to meet changes in procurement quantities. Simultaneously, it solves both the problems of consumption limit assessment and its fluctuations.

[0015] Furthermore, the number of exponential smoothing operations is determined based on the fluctuations in the historical consumption data. Using historical data on equipment and supplies to determine the number of exponential smoothing operations ensures the accuracy of the smoothing algorithm.

[0016] Furthermore, for cases where the historical fluctuation data is within a preset range (the preset range is generally the maximum absolute value of the historical fluctuation data), it indicates that the consumption rate of equipment is relatively stable. In this case, the purpose of estimating the equipment can be achieved by using one exponential smoothing method. However, for cases where the historical consumption data fluctuates greatly, it indicates that the consumption of equipment is highly volatile. In this case, the number of exponential smoothing operations is determined again to ensure the accuracy of the consumption data assessment.

[0017] Furthermore, for cases where the annual consumption data of equipment and supplies shows a unidirectional increase or decrease, it indicates that the fluctuation of consumed items is not within the preset range. For this phenomenon, the data can be determined using the quadratic exponential smoothing method. For cases where the corresponding data shows a curve-like upward and downward trend, the data can be determined using the triple exponential smoothing method. The number of exponential smoothing operations is determined according to different situations. On the one hand, this ensures the accuracy of data processing, and on the other hand, it ensures the speed of data calculation for regular data.

[0018] Furthermore, this method typically selects finalized products for calculation, and determines the initial value for the exponential smoothing method by combining the cumulative data years of equipment upgrades and replacements. This increases the accuracy of consumption limit assessment for equipment and the adaptability to assessment of similar equipment. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the method for calculating the consumption limit of equipment in the embodiments. Figure 2 This is a schematic diagram illustrating the process of determining the number of exponential smoothing operations in the embodiment. Detailed Implementation

[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0021] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0022] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0023] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] Please see Figures 1-2 As shown, Figure 1 This is a flowchart illustrating the method for calculating the consumption limit of equipment in the embodiments. Figure 2 This is a schematic diagram illustrating the process of determining the number of exponential smoothing operations in the embodiment.

[0025] This invention provides a method for calculating the consumption limit of equipment, including: Obtain historical consumption data and historical consumption estimates for any equipment category whose consumption needs to be evaluated. The number of exponential smoothing operations is determined based on the changes in consumption data over the years. Based on the determined number of exponential smoothing operations, the consumption of equipment is calculated using the exponential smoothing method. Historical fluctuation data is calculated by the difference between historical consumption data and historical consumption estimates. Based on uncertainty theory, the deviation of the current year's equipment consumption is estimated, and the consumption is adjusted according to the deviation of the equipment consumption. Based on the historical fluctuation data, a reliable deviation range is constructed, with consumption as the central factor. The consumption limit is a set that includes the consumption amount and the confidence interval of the deviation.

[0026] The process of calculating the deviation of consumption based on historical fluctuation data and uncertainty theory includes: Based on the historical fluctuation data, the floating range of the reliability measure as the standard value is determined by taking the mean square value, rounding up, expanding the range, and judging whether the reliability measure is satisfied. The floating range is then used as the deviation reliability range.

[0027] Specifically, the revised consumption figures can serve as a reference for the quantity of equipment allocated to units; the reliability range of the deviation, determined based on historical fluctuation data and reliability measurements, can serve as the basis for procurement fluctuations. In this way, the estimation of consumption figures is not affected, and the practical problem of insufficient fluctuation space after consumption figures are fixed, making it difficult to meet changes in procurement quantities, is solved simultaneously.

[0028] Specifically, historical consumption data and historical consumption estimates are obtained, and the number of exponential smoothing operations is determined based on the fluctuations in the historical consumption data. Historical fluctuation data is determined based on the historical consumption data and historical consumption estimates.

[0029] Specifically, the number of exponential smoothing operations is determined based on the fluctuations in the historical consumption data. Using historical data on equipment and supplies to determine the number of exponential smoothing operations ensures the accuracy of the smoothing algorithm.

[0030] Specifically, if the fluctuation of the annual consumption data is within a preset range, that is, the deviation between adjacent values ​​of the annual consumption data is within the maximum value of the absolute value of the annual fluctuation data, then the fluctuation is judged to be gentle and a single exponential smoothing method is adopted. The adjacent value deviation is the difference between the historical fluctuation data and the historical fluctuation data of adjacent years.

[0031] If the fluctuations in the annual consumption data are not within the preset range, the number of exponential smoothing operations will be determined based on the changing trend of the annual consumption data.

[0032] Specifically, if the fluctuations in historical consumption data are within a preset range, the purpose of estimating the consumption of items can be achieved by using one exponential smoothing method. However, if the fluctuations in historical consumption data are large, it indicates that the consumption of items is highly volatile. In this case, the number of exponential smoothing operations is determined again to ensure the accuracy of the consumption data assessment.

[0033] Specifically, for consumption data whose fluctuations over the years do not fall within the preset range, When historical consumption data shows a unidirectional increasing or decreasing trend, a quadratic exponential smoothing method is used. When the annual consumption data shows an upward or downward trend, the triple exponential smoothing method is used.

[0034] In practice, abnormal data generated by special reasons can be deleted during actual calculations. Simultaneously, a preliminary judgment can be made by combining the traditional exponential smoothing method with the presence or absence of trend and seasonality. That is, for no trend and no seasonality, a single exponential smoothing method is chosen; for trend but no seasonality, a double exponential smoothing method is used; and for trend and seasonality, a triple exponential smoothing method is used.

[0035] The exponential smoothing order mentioned in this invention is also referred to as the "order" in some other literature.

[0036] During implementation, a line graph is drawn with historical periods as the horizontal axis and annual consumption data as the vertical axis. The trend of annual consumption data is determined based on whether the distribution of points shows a linear or curvilinear trend.

[0037] Specifically, for equipment consumption data showing a unidirectional increase or decrease over the years, it indicates that although the fluctuations in consumption are not within the preset range, there is a certain pattern to the consumption data. For this phenomenon, a quadratic exponential smoothing method can be used to determine the data. For data with a curve showing an upward and downward trend, a triple exponential smoothing method can be used to determine the data. The number of exponential smoothing operations is determined according to different situations. On the one hand, this ensures the accuracy of data processing, and on the other hand, it ensures the speed of data calculation for regular data.

[0038] Regarding the determination of the initial value for the exponential smoothing method, considering the upgrading and replacement of equipment, the finalized product (the product in this manual refers to equipment) is generally selected for calculation. If the cumulative data after the product is finalized is greater than 10 years, the average value of the annual data of the product 10 years before the finalization is selected as the initial value for the exponential smoothing method. The annual consumption data of the equipment in the aforementioned equipment category within 10 years is iteratively processed based on the exponential smoothing method. If the product's cumulative data is less than or equal to 10 years, especially when the product has just been iterated and upgraded and the cumulative data is insufficient, the historical average of similar equipment before the upgrade can be used as the initial value for calculation. However, it is necessary to make weighted adjustments based on the product's functions and performance. For example, if the existing equipment can replace two existing ones, then the number of new products should be half of the original number. If there is no similar equipment, the initial value should be estimated using the expert scoring method.

[0039] Specifically, this method typically selects products that have been finalized for calculation. It combines the upgrading and replacement of equipment and the cumulative years of finalized data to determine the initial value for the exponential smoothing method, thereby increasing the accuracy of consumption limit assessment for equipment and the adaptability to different types of products.

[0040] Specifically, by subtracting any historical consumption data from the corresponding historical consumption estimate, the historical fluctuation data is calculated, and the deviation of consumption is calculated accordingly. For any equipment whose consumption is to be evaluated, set the estimated consumption limit as A, where A = P + R, and P is the corrected consumption amount and R is the deviation confidence interval.

[0041] (a) Calculate the consumption Step S1: Collect historical consumption data for a specific type of equipment. and historical consumption estimates ; and based on Calculate historical fluctuation data ; Step S2: Determine the number of times the exponential smoothing method will be used. The first exponential smoothing method is suitable for the consumption of a certain type of equipment that fluctuates but has no obvious trend over the years. The second exponential smoothing method is suitable for the consumption of a certain type of equipment that shows an upward or downward trend over the years. The third exponential smoothing method is suitable for the consumption of a certain type of equipment that shows a curved upward or downward trend over the years.

[0042] Step S3: If the product's cumulative data is less than or equal to 10 years, especially when the product has just been iterated and upgraded and the cumulative data is insufficient, the historical average of similar equipment before the upgrade can be used as an initial value for reference. However, it is necessary to make weighted adjustments according to the product's functions and performance. For example, if the existing equipment can replace two, then the new consumption amount is half of the original amount.

[0043] If the period (number of years) of the historical consumption data is ≥10, then the average of the historical consumption data from 10 years ago can be used as Y1; calculated using the exponential smoothing method with a step size of 0.1. Where Y1 is the first observation value, For smoothing coefficients; Specifically, the basic formula of the exponential smoothing formula In the formula, This is the predicted value for the next period. These are the actual observations from the current period. This is the predicted value for the current period; During implementation, historical consumption data for a certain type of equipment was collected. for Historical consumption estimates of a certain type of equipment for .

[0044] Step S4: Based on the exponential smoothing method and the calculation in step S3... Predict the consumption of a certain type of equipment to determine its consumption volume. ; (ii) Calculate the deviation value of consumption According to statistical theory, the uncertain distribution of a continuous random variable has one and only one mode. Based on the information obtained from the finite sample data, in The maximum, minimum, and mode of a sample sequence are determined by the following symbols: , , , ; Therefore, a triangular fuzzy variable is constructed. ; Introducing the fuzzy discretization method to handle triangular fuzzy variables Discretization processing. The triangular fuzzy variables are discretized using a fuzzy simulation discretization method (FS strategy). The support range is divided into Sub-intervals (in actual predictions, discretization can also be performed based on the actual deviation values, which makes it easier to understand the reliability of each deviation value): , , , ; Formation method: triangular fuzzy variables support range Total length is The left endpoint of the support interval is The right endpoint is This support range Average score If there are n subintervals, then the length of each subinterval is 1. Thus forming the above sub- Sub-intervals.

[0045] Furthermore, construct set functions. That is, the domain of the function is the set The form is that the probability of the interval is the maximum probability of all possible values ​​within that interval. Applying this set function to the above... Each sub-interval forms Number of values As shown below: , , , ; Right now , .

[0046] Constructing discrete fuzzy variables To replace triangular fuzzy variables Constructing discrete fuzzy variables It contains discrete values Discrete fuzzy variables The distribution is as follows: , in That is, to let discrete fuzzy variables Take discrete values The maximum value in; Based on triangular fuzzy variables Fuzzy membership function: , The calculation shows that: , Furthermore, calculate discrete fuzzy variables. Credibility distribution , , In summary, discrete fuzzy variables The credibility distribution is as follows: , And there are , in .

[0047] Note the discrete fuzzy variables generated by the above discretization methods. It satisfies the following convergence form: To ensure fuzzy variables It can be determined by discrete fuzzy variables. replace); Based on credibility measures and discrete fuzzy variables Calculate discrete fuzzy variables The equivalent value (real value): ; The above discrete fuzzy variables The equivalent value can approximate the triangular fuzzy variable. And determine it as the deviation of consumption.

[0048] (III) Calculate the confidence interval of the deviation Based on uncertainty theory, the specific method for calculating the deviation confidence interval R of the confidence measure as a standard value is as follows: Step S01: For any given year, the consumption of a certain type of equipment is NEy,y=1,2,...,T; Let NCy = NEy - NPy, then we obtain the historical fluctuation data of the difference between the annual consumption NEy and the corresponding annual forecast NPy. NCy is the difference between the consumption and the forecast value in any given year; Step S02, according to The numerical value is determined based on uncertainty theory as the deviation confidence interval R of the standard numerical value, and the standard numerical value mentioned in the implementation is 0.95; The specific calculation method for the aforementioned deviation confidence interval R is as follows: Step Sa is determined according to the following formula. That is, calculating the root mean square of the historical fluctuation data.

[0049] Step Sb, for Rounding up gives A, and then the floating range [-A, A] is obtained; Step Sc, Calculation The number M values ​​that do not belong to the floating range [-A, A]; Step Sd, Calculation ; Step Se, if Then the floating range is expanded to [-A-1, A+1]; Step Sf, repeat steps Sc, Sd, and Se, until the deviation value falling within the floating range is greater than 95%; At this point, let the floating space be the deviation confidence interval R.

[0050] Example 1

[0051] Since calculating consumption and the confidence interval of deviation is relatively simple, the example provides the calculation process for consumption deviation, specifically: Step 1: Based on existing historical fluctuation data, form sequence, The maximum value in the sequence is 4, the minimum value is 1, and the mode is 3. , , , Thus, a triangular fuzzy variable is constructed. .

[0052] Step 2: Discretize the triangular fuzzy variables using the fuzzy simulation discretization method (FS strategy). The support interval is divided into 5 sub-intervals (taken as a fixed interval). ): .

[0053] The length of each interval is Set functions Applying the above 5 sub-intervals will generate 5 values. As shown below: ; Step 3, construct discrete fuzzy variables set functions The credibility distribution of the above 5 sub-intervals, since the probability is uniformly distributed in this embodiment, corresponds to the 5 discrete values. They are respectively: .

[0054] Discrete fuzzy variables The credibility distribution is as follows: ; therefore, .

[0055] Step 4, based on credibility measures and discrete fuzzy variables Calculate discrete fuzzy variables Equivalent value (specific numerical value): , right Rounding up to 4, the deviation is 4. Step 5, Corrected consumption: Where P is the consumption predicted based on the exponential smoothing method, and 4 is the deviation of the consumption.

[0056] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0057] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calculating the consumption limit of equipment, characterized in that, include, Obtain historical consumption data and historical consumption estimates for any equipment category whose consumption needs to be evaluated. The number of exponential smoothing operations is determined based on the changes in consumption data over the years. Based on the determined number of exponential smoothing operations, the consumption of equipment is calculated using the exponential smoothing method. Historical fluctuation data is calculated by the difference between historical consumption data and historical consumption estimates. Based on uncertainty theory, the deviation of the current year's equipment consumption is estimated, and the consumption is adjusted according to the deviation. Based on historical fluctuation data, a reliable range of deviations is constructed with consumption as the core. The consumption limit is a set that includes the consumption amount and the confidence interval of the deviation.

2. The method for calculating the consumption of equipment as described in claim 1, characterized in that, The consumption of equipment was calculated using the exponential smoothing method based on historical consumption data.

3. The method for calculating the consumption of equipment as described in claim 2, characterized in that, The number of exponential smoothing operations is determined based on the fluctuations in consumption data over the years. If the fluctuations in the annual consumption data are within a preset range, that is, the deviation between adjacent values ​​of the annual consumption data is within the maximum value of the absolute value of the annual fluctuation data, then the fluctuation is determined to be gentle and the first exponential smoothing method is adopted. If the fluctuations in the annual consumption data are not within the preset range, the number of exponential smoothing operations will be determined based on the changing trend of the annual consumption data. If the historical consumption data shows a unidirectional increasing trend or a unidirectional decreasing trend, the quadratic exponential smoothing method should be used; If the consumption data over the years shows a curvilinear trend, the triple exponential smoothing method should be used.

4. The method for calculating the consumption of equipment as described in claim 2, characterized in that, The process of determining the initial values ​​for exponential smoothing includes: If the cumulative data period after the equipment type is finalized exceeds ten years, the average value of the annual consumption data of the equipment type ten years ago is selected as the initial value for the exponential smoothing method, and the annual consumption data of the equipment type within ten years is iteratively calculated based on the exponential smoothing method. If the cumulative data period of the equipment category is less than or equal to ten years, the historical average of similar equipment categories can be used as the initial value for calculation, and the historical average can be weighted and adjusted according to the differences in the characteristics of the selected equipment category. If there is no similar equipment, the initial value is estimated using the expert scoring method.

5. The method for calculating the consumption limit of equipment as described in claim 1, characterized in that, The process of estimating the deviation of equipment consumption for this year based on uncertainty theory includes: The deviation in consumption is calculated based on historical fluctuation data and uncertainty theory. The historical fluctuation data is the difference between the historical consumption data and the corresponding historical consumption estimate.

6. The method for calculating the consumption limit of equipment as described in claim 5, characterized in that, The process of calculating the deviation of equipment consumption for the current year based on historical fluctuation data includes: ① Extract the maximum, minimum, and mode of the historical fluctuation data; ② Construct a triangular fuzzy variable based on the maximum deviation of the maximum and minimum values ​​of the historical fluctuation data from the mode; ③ Based on the fuzzy discretization method, the triangular fuzzy variables are discretized and the support interval is divided into several sub-intervals. In the prediction of the deviation of equipment consumption, the historical fluctuation data is used as the deviation value to divide the sub-intervals, and the reliability measure of each deviation value is determined according to the probability of the deviation value. ④ Take the maximum value and probability maximum value of the discrete fuzzy variable in each support interval to determine the discrete fuzzy variable and its credibility measure, and determine the equivalent value of the discrete fuzzy variable; ⑤ Use the equivalent value of discrete fuzzy variables as the deviation of consumption.

7. The method for calculating the consumption limit of equipment as described in claim 1, characterized in that, The process of determining the confidence interval of the deviation of the consumption limit that meets the confidence measure as a standard value is as follows: ① Calculate the root mean square of the historical fluctuation data; ②Round up the root mean square of the historical fluctuation data; ③ Construct a floating range based on the floor function; ④ Calculate the proportion of historical fluctuation data that does not fall within the fluctuation range; ⑤ If the proportion of historical fluctuation data that does not belong to the floating range does not meet the set reliability measurement standard value range, then the range is expanded by one unit until the proportion of historical fluctuation data that does not belong to the floating range is within the preset reliability measurement standard value range, then the floating range is determined as the deviation reliability range.

Images