Aluminum electrolysis cell health state assessment method based on grey correlation analysis

Abstract

The invention relates to the technical field of aluminum electrolysis cell health state evaluation, in particular to an aluminum electrolysis cell health state evaluation method based on grey correlation analysis, which comprises the following steps: analyzing aluminum electrolysis cell state data to establish an aluminum electrolysis cell state evaluation index system; determining a subjective weight and an objective weight of each index in the aluminum electrolysis cell state evaluation index system, and combining the subjective weight and the objective weight; constructing a double-layer evaluation model based on the combined weight of each index, and solving the relative close degree and grey relational degree of each index based on the double-layer evaluation model; and combining the relative closeness with the grey correlation degree, solving the comprehensive state closeness, and completing the health state evaluation of the aluminum electrolysis cell based on the comprehensive state closeness. According to the invention, the state of the electrolytic cell can be comprehensively evaluated from different dimensions, and the accuracy and reliability of the evaluation result are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health status assessment of aluminum electrolysis cells, and in particular to a health status assessment method of aluminum electrolysis cells based on grey correlation analysis. Background Art

[0002] Aluminum electrolytic cells are the core equipment in aluminum industry production, and their operating status directly affects product quality and production efficiency. With the development of industrial automation and informatization, the operation monitoring system of aluminum electrolytic cells has been continuously improved, and a large amount of status data has been accumulated, which includes multiple parameters such as electrolyte level, electrolysis temperature, and cell working voltage during the operation of aluminum electrolytic cells, which contains rich status information. Through the analysis and processing of these status data of aluminum electrolytic cells, a scientific status assessment system can be established to achieve accurate assessment of the health status of aluminum electrolytic cells. At present, the traditional status assessment of aluminum electrolytic cells often only focuses on a single or a few indicators, such as voltage fluctuations or temperature changes, and ignores the correlation between indicators, resulting in one-sided assessment results and unable to fully reflect the true status of the electrolytic cell. At the same time, the selection of indicators lacks systematic and scientific basis, making it difficult to establish a complete assessment system; on the other hand, the traditional status assessment of aluminum electrolytic cells mostly adopts a single weight determination method, which either relies entirely on expert experience and is highly subjective; or only considers data characteristics and ignores the important value of expert experience. Based on this, in response to the above problems, we designed an aluminum electrolytic cell health status assessment method based on grey correlation analysis. Summary of the invention

[0003] The purpose of the present invention is to provide an aluminum electrolytic cell health status assessment method based on grey correlation analysis, which not only establishes a multi-level and multi-dimensional assessment index system by analyzing the electrolytic cell status data, but also innovatively introduces the game theory combined weighting method to organically combine the subjective weight with the objective weight, making the weight determination more scientific and reasonable; and constructs a two-layer evaluation model based on TOPSIS and grey correlation analysis, and comprehensively evaluates the electrolytic cell status from different dimensions by calculating the relative closeness and grey correlation, thereby improving the accuracy and reliability of the evaluation results.

[0004] The present invention is achieved through the following technical solutions:

[0005] A method for evaluating the health status of an aluminum electrolytic cell based on grey correlation analysis, the method comprising the following steps:

[0006] Acquire the status data of the aluminum electrolysis cell, analyze the status data of the aluminum electrolysis cell, obtain the aluminum electrolysis cell status evaluation index, and establish the aluminum electrolysis cell status evaluation index system based on the aluminum electrolysis cell status evaluation index;

[0007] Determine the subjective weight and objective weight of each indicator in the aluminum electrolytic cell status evaluation index system, and introduce the game theory combination weighting method to combine the subjective weight and objective weight of each indicator to obtain the combined weight of each indicator;

[0008] Based on the combined weights of each indicator, a two-layer evaluation model is constructed. The first layer model evaluates each indicator through the TOPSIS algorithm, and calculates the distance between each evaluation result and the optimal and worst ideal state to obtain the relative closeness of each evaluation result. The second layer model evaluates each indicator through grey correlation analysis, and calculates the grey correlation between each evaluation result and the optimal and worst ideal state indicators.

[0009] The relative closeness is combined with the grey correlation degree to solve the comprehensive state closeness, and the health status evaluation of the aluminum electrolytic cell is completed based on the comprehensive state closeness.

[0010] Optionally, before determining the subjective weight and objective weight of each indicator in the aluminum electrolysis cell state evaluation index system, it also includes: preprocessing the aluminum electrolysis cell state evaluation index system, and the preprocessing includes: positive processing and standardization processing.

[0011] Optionally, the subjective weight and objective weight of each indicator in the aluminum electrolytic cell state evaluation index system are determined, wherein the subjective weight of each indicator in the aluminum electrolytic cell state evaluation index system is determined specifically by a sequence relationship analysis method, which is specifically:

[0012] Rank the importance of each indicator in the aluminum electrolytic cell status evaluation indicator system;

[0013] Determine the importance between adjacent indicators;

[0014] Based on the importance ranking of each indicator and the importance degree between adjacent indicators, the weight coefficient of each indicator is obtained through recursive calculation;

[0015] The weight coefficients of the various indicators are combined to represent the subjective weights of the various indicators.

[0016] Optionally, the objective weight of each indicator in the aluminum electrolytic cell state evaluation index system is determined by an entropy weight method, which is specifically:

[0017] Based on the standardized aluminum electrolytic cell status evaluation index system, determine the standardized matrix of each index;

[0018] Convert the standardized matrix into the relative weight of each indicator, and calculate the entropy value of each indicator based on the information entropy theory;

[0019] Based on the entropy value of each indicator, the objective weight of each indicator is calculated.

[0020] Optionally, the objective weight of each indicator in the aluminum electrolysis cell state evaluation index system is determined by the following calculation formula:

[0021]

[0022] Among them, the standardized matrix Z = (z ij ) m×n , p ij For element z ij The proportion of the entire matrix, n is the total number of indicators, k = 1 / log(n), e j is the entropy of the jth index, w j is the objective weight, and m is the total number of samples.

[0023] Optionally, the game theory combined weighting method is introduced to combine the subjective weight and objective weight of each indicator, which is specifically:

[0024] The subjective weight and objective weight of each indicator are linearly combined to form a weight vector set;

[0025] Optimize the linear combination coefficients to obtain the optimal linear combination coefficients;

[0026] The optimal linear combination coefficients are normalized to obtain the combined weights of each indicator.

[0027] Optionally, the distance between each evaluation result and the optimal and worst ideal states is calculated as follows:

[0028]

[0029] Among them, d i + is the distance from the result to be evaluated to the optimal ideal state, d i - is the distance from the result to be evaluated to the worst ideal state.

[0030] Optionally, the grey correlation between each evaluation result and the best and worst ideal state indicators is calculated as follows:

[0031]

[0032]

[0033] Among them, r ij + It represents the grey correlation coefficient between the i-th group of aluminum electrolytic cell samples and the j-th index of the optimal ideal solution, r ij -represents the grey correlation coefficient between the i-th group of aluminum electrolytic cell samples and the j-th index of the worst ideal solution, ρ is the resolution coefficient, l i + represents the grey correlation between the i-th group of aluminum electrolytic cell samples and the optimal ideal solution, l i - It represents the grey correlation degree between the i-th group of aluminum electrolysis cell samples and the worst ideal solution.

[0034] Optionally, the specific process of solving the comprehensive state closeness is as follows:

[0035] The relative closeness and grey correlation are processed dimensionlessly;

[0036]

[0037]

[0038] The dimensionless relative closeness and grey correlation degree are weightedly fused;

[0039] C i + =αD i + +βL i + i∈n

[0040] C i - =αD i - +βL i - i∈n

[0041] Determine the comprehensive state closeness based on the weighted fusion result:

[0042]

[0043] Among them, D i + is the dimensionless d i + , D i - is the dimensionless d i - , L i + is the dimensionless l i + , L i - is the dimensionless l i - , C i +is the first weighted fusion result, C i - is the second weighted fusion result, α and β are preference coefficients respectively, E i It is the comprehensive state closeness.

[0044] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0045] The present invention not only establishes a multi-level and multi-dimensional evaluation index system by analyzing the electrolytic cell status data, but also innovatively introduces the game theory combined weighting method to organically combine the subjective weight with the objective weight, making the weight determination more scientific and reasonable; and constructs a two-layer evaluation model based on TOPSIS and grey correlation analysis, which comprehensively evaluates the electrolytic cell status from different dimensions by calculating the relative closeness and grey correlation, thereby improving the accuracy and reliability of the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of an electrolytic cell health status assessment index system provided by the present invention;

[0047] Figure 2 A schematic flow chart of the method for assessing the health status of an aluminum electrolysis cell based on grey correlation analysis provided by the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, what is described is only a part of the present invention, not all of it. Generally, the components of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0049] like Figure 1 , Figure 2 As shown, the present invention provides one embodiment: a method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis, the method comprising the following steps:

[0050] Acquire the status data of the aluminum electrolysis cell, analyze the status data of the aluminum electrolysis cell, obtain the aluminum electrolysis cell status evaluation index, and establish the aluminum electrolysis cell status evaluation index system based on the aluminum electrolysis cell status evaluation index;

[0051] Determine the subjective weight and objective weight of each indicator in the aluminum electrolytic cell status evaluation index system, and introduce the game theory combination weighting method to combine the subjective weight and objective weight of each indicator to obtain the combined weight of each indicator;

[0052] Based on the combined weights of each indicator, a two-layer evaluation model is constructed. The first layer model evaluates each indicator through the TOPSIS algorithm, and calculates the distance between each evaluation result and the optimal and worst ideal state to obtain the relative closeness of each evaluation result. The second layer model evaluates each indicator through grey correlation analysis, and calculates the grey correlation between each evaluation result and the optimal and worst ideal state indicators.

[0053] The relative closeness is combined with the grey correlation degree to solve the comprehensive state closeness, and the health status evaluation of the aluminum electrolytic cell is completed based on the comprehensive state closeness.

[0054] In this embodiment, this embodiment first analyzes and selects the state evaluation index of the electrolytic cell based on the operation mechanism and actual situation of the electrolytic cell by analyzing the state information of the electrolytic cell, and establishes an electrolytic cell state evaluation index system. On the basis of the index data, the data is preprocessed, first the data is positively processed, then the data is standardized, and finally the overall health status level of the electrolytic cell is reasonably divided. Secondly, according to the different degrees of influence of each evaluation index on the overall health status of the electrolytic cell, an electrolytic cell state evaluation index weight model is established. The subjective weight of each indicator is determined by the order relationship analysis method, the objective weight of each indicator is determined by the entropy weight method, and on the basis of a single weight, the game theory combination weighting method is introduced to determine the combined weight of each indicator. Finally, TOPSIS and gray correlation analysis are applied to the electrolytic cell state evaluation to establish a two-layer evaluation model. First, the first layer of the model uses TOPSIS to evaluate each indicator and calculates the distance between each evaluation sample and the optimal and worst ideal state. Secondly, the second layer of the model uses gray correlation analysis to evaluate each indicator and calculates the gray correlation between each state evaluation indicator and the optimal and worst ideal state indicators. Finally, the comprehensive state closeness was calculated by combining the results of TOPSIS and grey correlation analysis, and the electrolytic cell samples were comprehensively evaluated based on the comprehensive state closeness.

[0055] like Figure 1As shown, aluminum electrolysis is a complex nonlinear electrochemical reaction system with multivariable coupling. If the production management is not in place, it is easy to produce anode effect, aluminum liquid fluctuation, hot tank, cold tank and other faults or sick tank characteristics. The occurrence of these phenomena will lead to a decrease in output and an increase in energy consumption, or even a safety accident. These situations should be avoided as much as possible in normal production. These tank condition characteristics are related to the aluminum electrolysis process parameters, mainly including tank voltage, electrolyte temperature, electrolyte level, aluminum level, molecular ratio, pole distance, aluminum oxide concentration, anode effect coefficient, etc. The state of different parameters has different effects on the electrolytic cell. When the values ​​of these parameters are within the normal range, the electrolytic cell is in a normal operating state. When the values ​​of these parameters are in an abnormal state, the electrolytic cell may be in an abnormal or faulty state. Therefore, the present invention selects the eight process parameters of tank voltage, electrolyte temperature, electrolyte level, aluminum level, molecular ratio, pole distance, aluminum oxide concentration, and anode effect coefficient as indicators to characterize the health state of the electrolytic cell, thereby establishing an electrolytic cell state evaluation index system.

[0056] In further implementation, in order to facilitate the quantitative processing of the detected indicator data, this embodiment introduces two methods, positive and normalization, to pre-process the data:

[0057] Data normalization. After determining the status evaluation index system, since there are many and complex factors affecting the health status of the electrolyzer, and the dimensions and magnitudes of the evaluation indicators are different, the indicators will have large differences when compared with each other. Therefore, the positive method is introduced to normalize the data of each indicator.

[0058] For negative indicators, use the following formula to make them positive.

[0059] y ij =max(x ij )-x ij

[0060] Where: x ij Indicates the data of a certain indicator collected, y ij It indicates the value obtained after the indicator data is positively processed.

[0061] For interval indicators, use the following formula to make them positive.

[0062]

[0063] Where: [a, b] is the optimal interval of the indicator, M = max{a-min{x ij},max{x ij}-b}.

[0064] 2) Data standardization. Suppose the indicator evaluation matrix after positive processing is Y = (yij ) m×n , and then use formula (3) to standardize the normalized data.

[0065]

[0066] Where: z ij is a standardized matrix.

[0067] Then we get the normalized matrix Z = (z ij ) m×n :

[0068]

[0069] In this embodiment, the subjective weight and objective weight of each indicator in the aluminum electrolytic cell state evaluation index system are determined, wherein the subjective weight of each indicator in the aluminum electrolytic cell state evaluation index system is determined specifically by a sequence relationship analysis method, which is specifically:

[0070] Rank the importance of each indicator in the aluminum electrolytic cell status evaluation indicator system;

[0071] Determine the importance between adjacent indicators;

[0072] Based on the importance ranking of each indicator and the importance degree between adjacent indicators, the weight coefficient of each indicator is obtained through recursive calculation;

[0073] The weight coefficients of the various indicators are combined to represent the subjective weights of the various indicators.

[0074] In the implementation of this embodiment, the order relationship is first determined. For a series of evaluation indicators, their importance is ranked. If the indicator X i The importance is greater than (or equal to) X j , then record it as X i >X j , then for a set of indicators, we can get a relationship, which is:

[0075]

[0076] In the formula, n is the total number of indicators.

[0077] Determine two adjacent indices X i With X i-1 The importance ratio between n It can be expressed as follows:

[0078]

[0079] In the formula, w n-1 ,w nThey are respectively i-1 With X i The weight coefficient, r n The value of can be determined from the following table.

[0080] Table 2 r n Assignment reference

[0081]

[0082] Calculate the weight of each indicator. The weight coefficient of each indicator can be calculated according to the following two formulas.

[0083]

[0084] w n-1 =r n w n

[0085] Therefore, the subjective weight determined by the ordinal relationship analysis method is w i =[w1,w2,…,w n ].

[0086] In this embodiment, the objective weight of each indicator in the aluminum electrolytic cell state evaluation index system is determined by the entropy weight method, which is specifically:

[0087] Based on the standardized aluminum electrolytic cell status evaluation index system, determine the standardized matrix of each index;

[0088] Convert the standardized matrix into the relative weight of each indicator, and calculate the entropy value of each indicator based on the information entropy theory;

[0089]

[0090] Based on the entropy value of each indicator, the objective weight of each indicator is calculated;

[0091]

[0092] Among them, the standardized matrix Z = (z ij ) m×n , p ij For element z ij The proportion of the entire matrix, n is the total number of indicators, k = 1 / log(n), e j is the entropy of the jth index, w j is the objective weight, and m is the total number of samples.

[0093] In this embodiment, the game theory combined weighting method is introduced to combine the subjective weight and objective weight of each indicator, which is specifically:

[0094] The subjective weight and objective weight of each indicator are linearly combined to form a weight vector set W = {w1, w2, ..., w n}, where n basic weight vectors are combined by any linear combination;

[0095]

[0096] Optimize the linear combination coefficients so that the deviation between the optimal combination weight and all basic weights is minimized, and the optimal linear combination coefficients are obtained;

[0097]

[0098] The optimal linear combination coefficients are normalized to obtain the combined weights of each indicator.

[0099]

[0100] Where: w* is the combined weight, is the normalized linear combination coefficient.

[0101] In the specific implementation of this embodiment, the TOPSIS method (Technique for Order Preference by Similarity to an Ideal Solution) can also be called the superior and inferior solution distance method, which is a commonly used and effective method in multi-objective comprehensive evaluation decision-making. In its calculation process, the best evaluation index and the worst evaluation index are first determined, and then the proximity between the finite evaluation index and the ideal state index is calculated, and the relative proximity of the ideal solution is used as the criterion for comprehensive evaluation. In the state evaluation, if the state value of each indicator reaches the optimal value, the state is a positive ideal state:

[0102] S + =(S1 + ,S2 + , …,S n + )

[0103] On the contrary, if the state value of each indicator reaches the worst, the state is a negative ideal state:

[0104] S - =(S1 - ,S2 - ,…,S n - )

[0105] According to the weighted normalized index evaluation matrix, the distance between each evaluation sample and the optimal and worst ideal state is calculated. The distance d between the sample to be evaluated and the optimal ideal state is i+ for:

[0106]

[0107] The distance d from the sample to be evaluated to the worst ideal state i - for:

[0108]

[0109] Furthermore, the main idea of ​​the grey correlation analysis method is to judge the degree of connection between these sequences based on the similarity of the geometric shapes between different sequence curves, that is, a method to weigh the degree of a certain relationship between indicators according to the degree of dependence of the development trends between the indicators (grey correlation). First, the discrete behavioral observations of system factors are converted into piecewise continuous broken lines through the method of linear interpolation in numerical analysis. Secondly, a grey correlation model is constructed based on the geometric characteristics of the broken lines. For the geometric shapes between different broken lines, the closer the shapes are, the greater the grey correlation between the corresponding sequences, and vice versa. Introducing the grey correlation analysis method into the state evaluation of aluminum electrolytic cells can well understand the change process of the state of the evaluated object relative to the normal state. The calculation process is as follows:

[0110] 1) Calculate the grey correlation coefficient between each state evaluation index and the optimal and worst ideal state index:

[0111]

[0112] In the formula, r ij + It represents the grey correlation coefficient between the i-th group of aluminum electrolytic cell samples and the j-th index of the optimal ideal solution, r ij - It represents the grey correlation coefficient between the i-th group of aluminum electrolytic cell samples and the j-th index of the worst ideal solution. Among them, ρ is the resolution coefficient, which is usually taken as 0.5 according to the experience.

[0113] 2) Calculate the grey correlation between each state evaluation index and the optimal and worst ideal state index:

[0114]

[0115] In this embodiment, the specific process of solving the comprehensive state closeness is as follows:

[0116] 1) Determine the indicator evaluation matrix.

[0117]

[0118] 2) Determine the normalization matrix.

[0119]

[0120] 3) Determine the combined weight W of each indicator * =(w1 * ,w2 * ,…,w n * ).

[0121] 4) Determine the weighted normalization matrix A = (a ij ) m×n .

[0122]

[0123] 5) Determine the optimal ideal state S of each indicator + =(S1 + , S2 + , …, S n + ) and the worst ideal state S - =(S1 - , S2 - , …, S n - ).

[0124] 6) Determine the Euclidean distance d from each aluminum electrolytic cell sample to the optimal ideal state i + and the Euclidean distance d of the worst ideal state i - .

[0125] 7) Determine the grey correlation coefficient r between each aluminum electrolytic cell sample and the optimal ideal state ij + and the grey correlation coefficient r of the worst ideal state ij - .

[0126] 8) Determine the grey correlation degree I between each aluminum electrolytic cell sample and the optimal and worst ideal state i + ,I i - .

[0127] 9) The Euclidean distance and grey correlation degree are dimensionless.

[0128]

[0129] 10) Perform weighted fusion of dimensionless Euclidean distance and grey correlation. For the optimal ideal state, if the calculated Euclidean distance is smaller and the grey correlation is larger, the evaluation sample is closer to the ideal state. On the contrary, if the Euclidean distance is larger and the grey correlation is smaller, the sample is further away from the ideal state. Therefore, after fusion, we can get:

[0130] C i + =αD i + +βL i + i∈n

[0131] C i - =αD i - +βL i - i∈n

[0132] In the formula, α and β are preference coefficients, α+β=1, and the values ​​of α and β are between 0 and 1. i It can comprehensively reflect the closeness of the aluminum electrolysis cell sample to the ideal solution. i + The larger the value, the better the health of the aluminum electrolytic cell. i - The smaller the value, the worse the health of the aluminum electrolysis cell.

[0133] 11) Determine the degree of closeness of the comprehensive state. By analyzing the degree of closeness of the comprehensive state, we can get the degree of closeness between the evaluation sample and the optimal and worst ideal solutions. The degree of closeness of the comprehensive state of the i-th aluminum electrolytic cell sample is:

[0134]

[0135] 12) The conclusion is drawn. For the result obtained by formula (32), the larger the value, the closer the state of the aluminum electrolytic cell is to the optimal ideal solution, that is, the better the health state; the smaller the value, the closer the state of the aluminum electrolytic cell is to the worst ideal solution, that is, the worse the health state. By analyzing the degree of closeness of the comprehensive state of each group of aluminum electrolytic cells, not only can the state level of each aluminum electrolytic cell be obtained, but also the relative advantages and disadvantages of the states of each group of aluminum electrolytic cells can be compared. In this way, the final evaluation result can be obtained.

[0136] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for evaluating the health status of aluminum electrolytic cells based on grey correlation analysis, characterized in that: The steps of the method include: Acquire the status data of the aluminum electrolysis cell, analyze the status data of the aluminum electrolysis cell, obtain the aluminum electrolysis cell status evaluation index, and establish the aluminum electrolysis cell status evaluation index system based on the aluminum electrolysis cell status evaluation index; Determine the subjective weight and objective weight of each indicator in the aluminum electrolytic cell status evaluation index system, and introduce the game theory combination weighting method to combine the subjective weight and objective weight of each indicator to obtain the combined weight of each indicator; Based on the combined weights of each indicator, a two-layer evaluation model is constructed. The first layer model evaluates each indicator through the TOPSIS algorithm, and calculates the distance between each evaluation result and the optimal and worst ideal state to obtain the relative closeness of each evaluation result. The second layer model evaluates each indicator through grey correlation analysis, and calculates the grey correlation between each evaluation result and the optimal and worst ideal state indicators. The relative closeness is combined with the grey correlation degree to solve the comprehensive state closeness, and the health status evaluation of the aluminum electrolytic cell is completed based on the comprehensive state closeness.

2. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to claim 1, characterized in that: Before determining the subjective weight and objective weight of each indicator in the aluminum electrolysis cell state evaluation index system, it also includes: preprocessing the aluminum electrolysis cell state evaluation index system, and the preprocessing includes: positive processing and standardization processing.

3. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to claim 2 is characterized in that: The subjective weight and objective weight of each indicator in the aluminum electrolytic cell state evaluation index system are determined, wherein the subjective weight of each indicator in the aluminum electrolytic cell state evaluation index system is determined specifically by a sequence relationship analysis method, which is specifically: Rank the importance of each indicator in the aluminum electrolytic cell status evaluation indicator system; Determine the importance between adjacent indicators; Based on the importance ranking of each indicator and the importance degree between adjacent indicators, the weight coefficient of each indicator is obtained through recursive calculation; The weight coefficients of the various indicators are combined to represent the subjective weights of the various indicators.

4. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to claim 3 is characterized in that: The objective weight of each indicator in the aluminum electrolytic cell state evaluation index system is determined by the entropy weight method, which is as follows: Based on the standardized aluminum electrolytic cell status evaluation index system, determine the standardized matrix of each index; Convert the standardized matrix into the relative weight of each indicator, and calculate the entropy value of each indicator based on the information entropy theory; Based on the entropy value of each indicator, the objective weight of each indicator is calculated.

5. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to claim 4 is characterized in that: The objective weight of each indicator in the aluminum electrolysis cell state evaluation index system is determined by the following calculation formula: Among them, the standardized matrix Z = (z ij )m×n,p ij For element z ij The proportion of the entire matrix, n is the total number of indicators, k = 1 / log(n), e j is the entropy of the jth index, w j is the objective weight, and m is the total number of samples.

6. The method for evaluating the health status of aluminum electrolysis cells based on grey correlation analysis according to claim 5 is characterized in that: The game theory combined weighting method is introduced to combine the subjective weight and objective weight of each indicator, which is specifically: The subjective weight and objective weight of each indicator are linearly combined to form a weight vector set; Optimize the linear combination coefficients to obtain the optimal linear combination coefficients; The optimal linear combination coefficients are normalized to obtain the combined weights of each indicator.

7. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to any one of claims 1 to 6, characterized in that: The distance between each evaluation result and the optimal and worst ideal states is calculated as follows: Among them, d i + is the distance from the result to be evaluated to the optimal ideal state, d i - is the distance from the result to be evaluated to the worst ideal state.

8. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to claim 7, characterized in that: The grey correlation between each evaluation result and the best and worst ideal state indicators is calculated as follows: Among them, r ij + It represents the grey correlation coefficient between the i-th group of aluminum electrolytic cell samples and the j-th index of the optimal ideal solution, r ij - represents the grey correlation coefficient between the i-th group of aluminum electrolytic cell samples and the j-th index of the worst ideal solution, ρ is the resolution coefficient, l i + represents the grey correlation between the i-th group of aluminum electrolytic cell samples and the optimal ideal solution, l i - It represents the grey correlation degree between the i-th group of aluminum electrolysis cell samples and the worst ideal solution.

9. The method for evaluating the health status of an aluminum electrolysis cell based on grey correlation analysis according to claim 8, characterized in that: The specific process of solving the comprehensive state closeness is as follows: The relative closeness and grey correlation are processed dimensionlessly; The dimensionless relative closeness and grey correlation degree are weightedly fused; C i + =αD i + +βL i + i∈n C i - =αD i - +βL i - i∈n Determine the comprehensive state closeness based on the weighted fusion result: Among them, D i+ is the dimensionless d i + , D i - is the dimensionless d i - , L i + is the dimensionless l i + , L i - is the dimensionless l i - , C i + is the first weighted fusion result, C i - is the second weighted fusion result, α and β are preference coefficients respectively, E i It is the comprehensive state closeness.

Images