Multi-source heterogeneous index fusion evaluation method for pollution hidden dangers of non-ferrous metal smelting enterprises

Abstract

The embodiment of the invention discloses a non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion assessment method, which comprises the following steps: acquiring a plurality of assessment indexes of non-ferrous metal smelting enterprise pollution hidden dangers and weights of the assessment indexes; determining membership degree distribution of each qualitative index under different grades according to the grade of each qualitative index in the plurality of evaluation indexes; and obtaining a comprehensive evaluation result of the enterprise according to the membership distribution and the weight of each evaluation index. According to the embodiment, multi-source heterogeneous indexes can be uniformly processed, and comprehensive evaluation of enterprise pollution hidden dangers is realized.

Description

TECHNICAL FIELD

[0001] The embodiment of the application relates to the technical field of environmental monitoring and evaluation, and particularly relates to a non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method. BACKGROUND

[0002] The non-ferrous metal smelting industry is a typical high-pollution and high-energy-consumption industry. The core processes such as smelting, roasting and electrolysis in the production process will emit a large amount of wastewater, waste gas, waste residue and pollutants containing heavy metals such as arsenic, cadmium, lead and zinc, which pose a serious threat to the ecological environment (especially soil and groundwater) and human health. Although the ecological environment department implements strict control on the total amount of pollutants discharged by enterprises, due to the existence of objective factors such as management defects, equipment aging and natural environmental influence, environmental pollution events or ecological damage risks are still difficult to completely avoid in the actual production process. Therefore, it is crucial to systematically identify and scientifically evaluate the conditions, behaviors or states (i.e. "pollution hidden dangers") that can induce environmental problems in the industry.

[0003] In the prior art, the most similar scheme to the present patent technology mainly includes the following two types: The first type is based on risk evaluation technology after pollution occurs. For example, the "scenario-based ecological risk evaluation method for contaminated sites" proposed in patent CN114139901A, which quantitatively evaluates the ecological risk of contaminated sites by constructing a pollution diffusion model and a receptor exposure model. This method mainly focuses on the migration and transformation process of pollutants in the environment and the exposure pathway of receptors, but lacks systematic analysis of the production link of the pollution source. Similarly, the "enterprise land soil and groundwater environmental risk identification method based on public data" developed by patent CN116245353A, although it introduces multi-source data fusion technology, its evaluation dimension is still limited to the identification and evaluation of pollution consequences.

[0004] The second type is the traditional fuzzy comprehensive evaluation technology. This kind of method usually adopts a single path when dealing with evaluation systems with fuzzy characteristics: only dealing with qualitative indicators such as equipment management level scored by experts; or only dealing with quantifiable quantitative indicators such as pollutant emission concentration. However, in actual application, enterprise pollution hidden danger evaluation needs to integrate multi-source heterogeneous indicators such as production process parameters, environmental monitoring data, equipment operating status and management system, which have significant differences in data types, dimensions and fuzzy characteristics, making it difficult for traditional methods to establish a unified evaluation dimension.

[0005] In the evaluation of the complex system of enterprise pollution hidden danger, qualitative indicators and quantitative indicators are highly intertwined and coexist, and the evaluation process deeply depends on the above-mentioned multi-source heterogeneous information. The mixed characteristics make it difficult to effectively compare, integrate and comprehensively calculate different types, different sources and different dimensions of indicators in the same evaluation system. How to uniformly process these multi-source heterogeneous indicators and realize the comprehensive evaluation of enterprise pollution hidden danger is a problem to be solved. SUMMARY

[0006] The embodiment of the present application provides a non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method to solve the above technical problems.

[0007] In the first aspect, the embodiment of the present application provides a non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method, comprising: Obtaining a plurality of evaluation indicators of non-ferrous metal smelting enterprise pollution hidden danger and the weight of each evaluation indicator; According to the grade of each qualitative indicator in the plurality of evaluation indicators, the membership degree distribution of each qualitative indicator in different grades is determined; According to the membership degree distribution and the weight of each evaluation indicator, the comprehensive evaluation result of the enterprise is obtained.

[0008] In the second aspect, the embodiment of the present application provides an electronic device, comprising: One or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors realize the non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method described in any embodiment.

[0009] In the third aspect, the embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method described in any embodiment.

[0010] The embodiment provides a non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method. In order to realize the fusion of qualitative indicators and quantitative indicators, the fuzzy evaluation is used for quantitative analysis of qualitative indicators, the membership degree distribution vector is designed, and the fuzzy mathematical principle is used to convert it into a calculable value, so as to solve the problem of subjective scoring in the traditional method, enhance the consistency and repeatability of evaluation. At the same time, the normalization rule of quantitative indicators and the membership function matching mechanism are proposed, the quantitative indicators are also converted into membership degree distribution vectors, the dimension difference between qualitative indicators and quantitative indicators is eliminated, the comprehensive calculation of heterogeneous data in the same evaluation system is realized, the data utilization rate is improved, and the objectivity and accuracy of evaluation are improved. Attached Figure Description

[0011] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a method for fusion assessment of multi-source heterogeneous indicators of pollution risks in non-ferrous metal smelting enterprises, provided by an embodiment of the present invention. Figure 2 This is a schematic diagram of a pollution hazard evaluation index system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a membership matrix provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0014] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0015] In the description of the present application, it should be further pointed out that, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be direct connection, or indirect connection through intermediate medium, or internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0016] As described in the background, the existing evaluation method has the following problems: ①Incomplete index coverage: the existing evaluation system often focuses on a single environmental medium or a specific type of pollutant, and fails to systematically cover various pollution hazards that may occur in the whole process of non-ferrous metal smelting. This is due to the lack of systematic collection, integration and management of environmental protection information related to enterprise production and management "whole chain" (from raw materials, production process, equipment operation, pollution control to environmental monitoring). These information sources come from multiple sources such as field investigation, production report, monitoring data, equipment account, etc., showing significant characteristics of multi-modal (text, numerical value, image), multi-source heterogeneous (different formats, standards), multi-disciplinary (involving environment, chemical industry, material, safety, etc.), multi-medium (involving water, gas, slag, soil, groundwater, etc.). This complexity directly leads to the partial absence of hidden danger evaluation index and the formation of data barrier.

[0017] ②Time lag: the existing evaluation method focuses on the consequence evaluation after pollution occurs, and lacks systematic identification of hidden factors before pollution occurs.

[0018] ③Single dimension: traditional fuzzy evaluation method cannot effectively integrate the subjective fuzziness of qualitative indicators and the objective fuzziness of quantitative indicators.

[0019] Figure 1 The flowchart of a non-ferrous metal smelting enterprise pollution hazard multi-source heterogeneous index fusion evaluation method provided by the embodiment of the present application is shown in the figure, to solve at least one of the above problems. The method is executed by an electronic device, as shown in the figure, the method specifically comprises: Figure 1 S110, obtaining a plurality of evaluation indexes of non-ferrous metal smelting enterprise pollution hazard, and the weight of each evaluation index. S110, obtaining a plurality of evaluation indexes of non-ferrous metal smelting enterprise pollution hazard, and the weight of each evaluation index.

[0020] In this embodiment, the evaluation data of the non-ferrous metal smelting enterprise pollution hazard to be evaluated is obtained, including the value of each evaluation index and the weight of each evaluation index, as the data source of the whole method.

[0021] In a specific embodiment, first, the evaluation system of non-ferrous metal smelting enterprise pollution hazard is constructed, which can include the following steps: S1: Evaluation index system construction.

[0022] S11: Determine the index system framework. Optionally, based on the formation mechanism and migration path of heavy metal pollution in non-ferrous metal smelting enterprises, the pollution hazard evaluation index can be divided into four categories: personnel operation, equipment state, management system, and environmental characteristics; using system analysis method, the index system is divided into three levels of target layer (A), criterion layer (C), and index layer (P), wherein the target layer is the comprehensive evaluation of pollution hazards in non-ferrous metal smelting enterprises, and the criterion layer includes personnel criterion (C1), equipment criterion (C2), management criterion (C3), and environmental criterion (C4).

[0023] S12: Evaluation index screening. Optionally, through the analysis of enterprise environmental risk at home and abroad, pollution site investigation, pollution hazard investigation cases, environmental protection supervision cases, and other aspects, according to the scientificity, intuitiveness, completeness and operability of index selection, combined with evaluation target, a number of evaluation indexes can be preliminarily screened from the index layer to obtain the original index system.

[0024] Further, the indexes of the personnel criterion can include: employee environmental safety awareness level, employee training frequency, employee environmental training effect, personnel operation skill proficiency, and illegal operation rate.

[0025] The indexes of the equipment criterion can include: equipment aging degree, equipment integrity, protective facility equipment condition, protective facility integrity, and hazardous chemical substance storage compliance.

[0026] The indexes of the management criterion can include: leakage environmental risk, wastewater environmental risk, waste gas environmental risk, solid waste environmental risk, and enterprise environmental illegal behavior times.

[0027] The indexes of the environmental criterion can include: regional temperature, regional rainfall, production temperature, material pH value, and factory impermeable surface coverage rate.

[0028] For example, the complete index system is shown in Figure 2 .

[0029] S13: Determine the evaluation set. Optionally, according to the actual needs of evaluation decision, the enterprise pollution hazard level can be divided into five levels: very low, low, medium, high, and very high. That is, the evaluation set V={V1, V2, V3, V4, V5}={very low, low, medium, high, very high}.

[0030] S2: Determine the weight of the element based on the analytic hierarchy process.

[0031] To scientifically quantify the relative importance degree of each element (personnel criterion, equipment criterion, management criterion, environment criterion) in the criterion layer and the specific index under each criterion layer to the target layer (comprehensive evaluation of pollution hazards of non-ferrous metal smelting enterprises), the embodiment adopts the analytic hierarchy process (AHP) to determine the weight coefficients of each level. The AHP method decomposes a complex problem into a hierarchical structure, and compares and judges the elements in the same level two by two, so as to realize the quantitative processing of qualitative problems, and finally calculates the weight value reflecting the relative importance of the elements. The specific implementation steps are as follows: S21: Construct a judgment matrix. Optionally, for each criterion layer and each index layer element in S12, the importance relative to the directly upper element is compared two by two. The 1-9 scale method (see Table 1) proposed by Saaty is used for assignment, and a 4*4 order judgment matrix M A and a 4*4 order judgment matrix M Ai are constructed, where i=1, 2, 3, 4; n is the number of indexes in the i-th layer.

[0032] Table 1 1-9 scale method

[0033] S22: Calculate the single-layer weight vector. Optionally, for each constructed judgment matrix, the maximum eigenvalue λ max and the corresponding normalized eigenvector W A and W Ai of the judgment matrix M A and M Ai are calculated respectively by using the eigenvector method. The eigenvector of the criterion layer to the target layer is W A =[w C1 ,w C2 ,w C3 ,w C4 ] T ; the eigenvector of the index layer to the criterion layer is W Ai =[w Ci1 ,w Ci2 ,...,w Cin ] T S23: Consistency check. Optionally, the following steps are included: a. Calculate the consistency index (CI): where n is the order of the matrix.

[0034] b. Find the average random consistency index (RI), which varies with the size of the matrix (n) (standard RI values table can be found, such as n=3, RI≈0.58, n=4, RI≈0.90, n=5, RI≈1.12).

[0035] c. Calculate the consistency ratio (CR): .

[0036] d. Determine: when CR<0.10, the judgment matrix is considered to have satisfactory consistency, and the calculated weight vector can be accepted. If CR≥0.10, it indicates that the consistency of the judgment matrix is poor, and the element assignment in the judgment matrix needs to be adjusted until the consistency test is passed.

[0037] S24: Determine the element weight subset. Optionally, through the calculation and inspection of steps S21-S23 of the judgment matrix M A and M Ai , the characteristic vectors of each layer are obtained, which respectively constitute the criterion layer weight subset A and the index layer weight subset A i : A={w C1 ,w C2 ,w C3 ,w C4} and satisfy ; A i ={w Ci1 ,w Ci2 ,...,w Cin} and satisfy ; The above evaluation index system and weight jointly constitute the evaluation system of pollution hazards of non-ferrous metal smelting enterprises. The embodiment can score or calculate specific values for each evaluation index of the enterprise to be evaluated under the evaluation system to obtain the above evaluation data.

[0038] S120, according to the grade of each qualitative index in the plurality of evaluation indexes, determine the membership degree distribution of each qualitative index under different grades.

[0039] This step constructs a multi-source heterogeneous index fusion evaluation matrix according to the specific data of each evaluation index obtained. Among them, there are qualitative indexes in the evaluation indexes, and there are also quantitative indexes. The evaluation results of qualitative indexes (such as the level of employee environmental safety awareness, employee environmental training effect, personnel operation skill proficiency, equipment aging degree, equipment integrity, protection facility integrity, and dangerous chemical storage compliance) are a certain level (such as a certain level in the above five levels). The evaluation results of quantitative indexes (such as employee training frequency, illegal operation rate, leakage environmental risk, wastewater environmental risk, waste gas environmental risk, solid waste environmental risk, enterprise environmental illegal behavior times, regional temperature, regional rainfall, production temperature, material acid-base degree, and factory impermeable surface coverage) are a certain specific value, such as: 1. Illegal operation rate = ( ) × 100% ) × 100% 2. The quantification methods of leakage environmental risk, wastewater environmental risk, solid waste environmental risk, and enterprise environmental illegal behavior times refer to the “Technical Regulations for Risk Screening and Risk Classification of In-Production Enterprises” (Trial).

[0040] 3. Factory impermeable surface coverage rate = ( ) × 100% ) × 100% This step converts the evaluation results of qualitative indexes and quantitative indexes into membership degree distribution vectors belonging to different levels of each index, and the fusion evaluation matrix is composed of these vectors.

[0041] In a specific embodiment, for qualitative indexes, the membership degree distribution of each qualitative index at different levels can be determined according to the level of each qualitative index in the plurality of evaluation indexes, that is, when a certain qualitative index is evaluated to a certain level, this step determines the probability (i.e. membership degree) that the qualitative index actually belongs to different levels. Optionally, the process can include the following steps: S31, in the case where each qualitative index is divided into five levels of very good, good, general, poor, and very poor (the specific number of levels can be adjusted as needed), a membership degree distribution vector r j = (r j1 , r j2 , …, r j5 ) can be designed for each of the five levels, where j represents the jth qualitative index, and r j1 -r j5 are five evaluation membership values of the jth qualitative index, i.e. the probability of the jth qualitative index belonging to five levels. Then, the conversion from the original evaluation result to the membership degree distribution vector can follow the following principles: U1, the center level has the highest membership degree: when the current qualitative index is evaluated as a certain level, the membership degree of the level is the highest, and the level is also called the center level.

[0042] U2, the adjacent level is non-zero: considering the boundary ambiguity, the level adjacent to the center level is given a non-zero value.

[0043] U3, the sum is 1: the membership vector of the same description needs to be normalized.

[0044] U4, special rules are customized for boundary scenarios to avoid evaluation conflicts. For example, in some special scenarios, the current qualitative index can only belong to a certain level, and then the membership degree of the level is forced to be 1 and the membership degrees of other levels are forced to be 0 through special rules.

[0045] Optionally, for each qualitative index in the multiple evaluation indexes: based on historical evaluation data and in combination with the following principles, the membership degree corresponding relationship of each level under the same qualitative index is determined, wherein the membership degree corresponding relationship is used to express the membership degrees of the same qualitative index in different levels when the qualitative index is evaluated as each level: Principle one: when a certain qualitative index is evaluated as a certain level, the membership degree of the certain qualitative index in the certain level is assigned as the highest membership degree, and the membership degrees of the certain qualitative index in the adjacent levels of the certain level are assigned as non-zero membership degrees (i.e., U1 and U2 above).

[0046] Principle two: when a certain qualitative index is evaluated as a certain level, the sum of the membership degrees of the certain qualitative index in different levels is 1 (i.e., U3 above), and U4.

[0047] In a specific embodiment, the membership degree corresponding relationship can be determined in the following manner: first, construct standard samples: collect a large number of "standard samples" representing different levels from historical data, or directly design such standard samples (such as standard videos, standard data, or simulated behaviors) according to specification requirements. Then, organize large-scale evaluation: let a large enough evaluator group (which can be much larger than the expert group) evaluate these "standard samples". The key point is that instead of directly grading them, they are asked to give membership degree scores (0-1) for each sample in each evaluation level. Then, perform statistical analysis: for all samples that are unanimously (such as more than 80%) identified as a certain level, calculate their average membership degrees in each level. Finally, in combination with the above principles one and two, correct each average membership degree to obtain the final membership degree corresponding relationship. This method gives the basic characteristics of each level by constructing standard samples, and then summarizes the fuzziness of each level through large-scale evaluation scoring, and the obtained membership degree relationship is more objective and reasonable than single-level evaluation (i.e., directly determining which level it belongs to).

[0048] Furthermore, the above membership degree correspondence can be expressed as follows: Figure 3 The membership matrix shown represents the probability (i.e., membership degree) that a qualitative indicator actually belongs to the level corresponding to a certain row when it is evaluated as "very good". For example, the first row indicates that when a qualitative indicator is evaluated as "very good", the probability of it belonging to "very good" is 0.9, the probability of it belonging to "good" is 0.1, and the probabilities of it belonging to "average", "poor", and "very poor" are all 0. Because the evaluation of qualitative indicators is subjective and ambiguous, when the evaluation result is "very good", the probability of it actually belonging to "very good" is the highest, but some may also belong to "good" (close to "very good"). Therefore, this embodiment includes both "very good" and "good" in the evaluation result and uses membership degrees to reflect the probability of each level, which is more conducive to reflecting the objective situation and distinguishing different levels. Each qualitative indicator corresponds to such a membership matrix.

[0049] After obtaining the above membership degree correspondence, we can convert the evaluation level of each qualitative indicator into a membership degree distribution of each qualitative indicator belonging to different levels based on this relationship.

[0050] In one specific implementation, for quantitative indicators: fuzzy evaluation methods can be used to convert the evaluation values ​​of each quantitative indicator into membership degree distributions of different levels to which each quantitative indicator belongs. Optionally, this process may include the following steps: S32: Normalization of Quantitative Indicators. Indicators are categorized into positive, negative, and interval indicators. For positive indicators, higher values ​​are better; for negative indicators, lower values ​​are better; and for interval indicators, the optimal value is within a specific interval.

[0051] Positive indicator normalization:

[0052] Negative indicator normalization:

[0053] Normalization of interval indicators:

[0054]

[0055] Where, x i It is a set of interval-type indicator sequences, with the optimal interval being [a, b].

[0056] S33: Selection Criteria for Membership Functions of Quantitative Indicators. This involves matching appropriate membership functions to the normalized numerical values, mapping them to the membership degrees of a fuzzy language set, and obtaining the evaluation vector for quantitative indicators, thus providing a foundation for fuzzy comprehensive evaluation.

[0057] Specifically, for the boundary clear indicators (such as the rate of irregular operation, the number of environmental violations), the triangular function can be selected to calculate the membership degree; for the interval type indicators (such as production temperature, material acidity and alkalinity), the trapezoidal function can be selected to calculate the membership degree. The calculation formula is as follows: Triangular function:

[0058] Trapezoidal function:

[0059] Wherein, a, b, c, d are all different levels of demarcation point. The quantitative index evaluation vector calculated by the membership function can be denoted as r k = (r k1 , r k2 ,..., r k5 ), wherein k is the kth quantitative index, and r k1 -r k5 are the five evaluation membership degrees of the kth qualitative index. More details of fuzzy evaluation are prior art, which will not be repeated here.

[0060] Finally, the index layer single element evaluation matrix is established by fusing the qualitative and quantitative indicators. Specifically: Through S31, the qualitative index evaluation vector (i.e. the membership distribution vector) is obtained as r j = (r j1 , r j2 ,..., r j5 ), through S32 and S33, the evaluation vector of the quantitative index is obtained as r k = (r k1 , r k2 ,..., r k5 ), and the evaluation decision matrix R i of the single element in each index layer is composed of the evaluation vectors of the qualitative index and the quantitative index, the evaluation set contains 5 levels, which can be represented as a matrix:

[0061] Wherein, i = 1, 2, 3, 4; n is the number of each criterion layer index.

[0062] S130, according to the membership distribution and weight of each evaluation index, the comprehensive evaluation result of the enterprise is obtained.

[0063] This step obtains the comprehensive evaluation result according to the comprehensive evaluation matrix. In a specific embodiment, the process can include the following steps: S41: Calculate the evaluation results of each index layer. Optionally, by using ordinary matrix multiplication, after synthesis operation, the comprehensive evaluation results of the personnel operation, equipment state, management system and environmental characteristics four criterion layers are obtained: B i =A i ·R i In the formula, B i are the evaluation result vectors of the personnel operation, equipment state, management system and environmental characteristics four criterion layers (i=1, 2, 3, 4); A i is the weight vector of the index layer (obtained from S24); R i is the evaluation matrix of the index layer (obtained from S34).

[0064] S42: Calculate the comprehensive evaluation matrix of the target layer. Optionally, R=(B1, B2, B3, B4) T , in the formula, R is the evaluation matrix of the target layer (enterprise pollution hidden danger), which is composed of four criterion layer evaluation result vectors (B1-B4, obtained from S41).

[0065] S43: Determine the fuzzy comprehensive evaluation result of the pollution hidden danger. Optionally, the fuzzy comprehensive evaluation result of the enterprise pollution hidden danger B=A·R, in the formula, B is the five-dimensional vector of the fuzzy comprehensive evaluation result of the enterprise pollution hidden danger, and the evaluation grade corresponding to the maximum value in the vector is the enterprise pollution hidden danger grade; A is the weight vector of the criterion layer (obtained from S24); R is the evaluation matrix of the criterion layer (obtained from S42).

[0066] In summary, the embodiment provides a non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method, which can realize the following beneficial effects: 1. The scientificity and precision of the evaluation system are improved. In order to overcome the defects of one-sidedness of the index in the prior art, the embodiment closely surrounds the formation mechanism and migration path of soil and groundwater pollution, and constructs a set of systematic index system throughout the whole process of pollution hidden danger generation, pollution migration and exposure. The key indexes in the core fields of personnel operation, equipment state, management system and environmental characteristics are deeply integrated, the whole process identification of pollution hidden danger is realized, and the hidden danger detection rate is significantly improved.

[0067] 2. In order to realize the fusion of qualitative indicators and quantitative indicators, the qualitative indicators are quantified by means of fuzzy evaluation, the five-level membership degree distribution vector is designed, and the five-level membership degree distribution vector is converted into a calculable value through fuzzy mathematical principles, so as to solve the problem of subjective scoring in traditional methods, enhance the consistency and repeatability of evaluation. At the same time, the normalization rule of quantitative indicators and the matching mechanism of membership degree function are proposed, so that the quantitative indicators are also converted into five-level membership degree distribution vectors, the dimensional difference between qualitative indicators and quantitative indicators is eliminated, the comprehensive calculation of heterogeneous data under the same evaluation system is realized, the data utilization rate is improved, and the objectivity and accuracy of evaluation are improved.

[0068] 3. It can accurately locate high-risk links (such as the combined risk of equipment aging and management loopholes), realize preventive maintenance, and greatly reduce the cost of handling sudden environmental events. The unified evaluation framework directly interfaces with the enterprise production system, reduces the manual data integration time, and improves the efficiency of environmental risk assessment.

[0069] 4. Through the pre-identification of hidden dangers (such as predicting pollution risks by combining material pH and impermeable area of the factory), it can reduce the emission of pollutants and reduce the long-term impact on soil and groundwater; at the same time, it can dynamically monitor environmental parameters (such as regional rainfall), adjust the prevention and control measures in time, and effectively curb the spread of pollution.

[0070] In a specific embodiment, there is also a situation that a large number of enterprises have the same qualitative indicators concentrated in one or a few levels, and it is difficult to distinguish the advantages and disadvantages of each enterprise. At this time, it is likely that there is a subjective bias in the evaluation of qualitative indicators (such as universal preference or universal bias), or the existing level standard deviates from the actual situation, and cannot reflect the differences between enterprises. This situation is particularly unfavorable in the case of sorting and distinguishing the level of the same qualitative indicators of multiple enterprises, and it is difficult to distinguish the differences between enterprises, and it loses the significance of evaluation benchmarking. In view of this situation, when determining the membership degree corresponding relationship of each level of the qualitative indicators, the uniformity of the number distribution of enterprises in each level is taken into account, so that the number distribution of enterprises in each level of the same qualitative indicators is uniform according to the membership degree corresponding relationship.

[0071] Optionally, for the same qualitative indicators: first, determine the number of enterprises evaluated in each level according to the evaluation results of each enterprise. If the number of enterprises in each level is very uneven, such as concentrated in one or two levels, the membership degree corresponding relationship (such as the membership degree matrix shown in FIG. 8) can be determined through mathematical programming method. Figure 3

[0072] ​Specifically, when the same qualitative indicator is evaluated to a certain level, this level is taken as the central level. The variables to be planned in the mathematical programming method are: the membership degree of the same qualitative indicator under each central level, and the membership degree under adjacent levels of each central level. The remaining membership degrees in the membership matrix are 0.

[0073] The constraints of mathematical programming are as follows: the more enterprises that are evaluated to a certain level for the same qualitative indicator, the lower the membership degree of the same qualitative indicator under the certain level as the central level; the membership degree of the central level is greater than the membership degree of the adjacent levels, the direction of change of the membership degree of the central level is opposite to the direction of change of the membership degree of the adjacent levels, and the step size of the change of the central level is greater than the step size of the change of each adjacent level; the ranking of the superiority and inferiority of the levels evaluated by different enterprises is consistent with the order of superiority and inferiority represented by the membership degree of different enterprises under the same level.

[0074] The goal of mathematical programming is to redetermine the number of enterprises belonging to each level based on the new membership degree of each enterprise under different levels in each mathematical programming operation; the optimal goal is achieved when the number of enterprises belonging to each level is evenly distributed.

[0075] The following is combined Figure 3 The mathematical programming method described above will be explained in detail. For example... Figure 3 As shown, when a certain qualitative indicator is evaluated as "very good," corresponding to "very good" in the row index of the membership matrix, then "very good" in the column index represents the central level. Matrix elements with both rows and columns marked as "very good" (i.e., diagonal elements) represent the membership levels under the central level, and "good" in the column index represents the adjacent levels of the central level. Therefore, in this embodiment's mathematical programming method, the variables to be programmed are the diagonal elements of the membership matrix, and the two adjacent elements in the same row as the diagonal elements corresponding to the three middle levels "good," "average," and "poor." Figure 3 The matrix elements are shown in the red box. Optionally, the membership degrees to be planned can be discretely selected, with the step size of each central membership level (i.e., the diagonal elements) being greater than the step size of other membership degrees to be planned. For example, each diagonal element can be discretely selected with a step size of 0.05, and the other adjacent elements in the same row as the diagonal elements can be discretely selected with a step size of 0.01. In each iteration of the mathematical programming operation, the objective function is calculated based on a selected combination of discrete degree values ​​that meets the constraints, until the objective function is optimal.

[0076] Optionally, the constraints of this mathematical programming method include: 1) In the previous step, the number of enterprises in each grade has been calculated according to the evaluation data, and the more the number of enterprises in a certain grade, the more concentrated the number of enterprises in the grade, which does not meet the needs of uniform evaluation, so these enterprises should be dispersed to adjacent grades to achieve differentiation. Therefore, the membership degree of the center grade can be planned lower to ensure that the membership degrees of adjacent grades are improved to achieve differentiation between the center grade and adjacent grades. For example, assuming that 90% of the enterprises in the evaluation result are concentrated in the "general" grade and 10% are concentrated in the "better" grade, the following constraints can be imposed on the diagonal elements in the membership matrix: the diagonal element a33 corresponding to "general" is the lowest, the diagonal element a22 corresponding to "better" is higher than a33, and the remaining three diagonal elements a11, a44 and a55 are all higher than a33. Figure 3

[0077] 2) In each planning operation, for the same row elements in the membership matrix, when the diagonal element increases relative to the last planning operation, the sum of the adjacent elements of the diagonal element must decrease relative to the last planning operation; similarly, when the diagonal element decreases relative to the last planning operation, the sum of the adjacent elements of the diagonal element must increase relative to the last planning operation; and the sum of all elements in the same row is 1.

[0078] 3) In each planning operation, the diagonal element is greater than the adjacent elements of the diagonal element in the same row elements of the membership matrix, i.e. the membership degree of the center grade is still the largest, thereby preserving the basic information in the original evaluation result.

[0079] 4) After converting the grades of different enterprises evaluated into membership distribution vectors under different registrations using the membership matrix in each planning operation, the order of the advantages and disadvantages represented by the membership degrees of different enterprises in the same grade is consistent with the order of the advantages and disadvantages of the original evaluation grades. For example, enterprises 1 and 2 are respectively evaluated as "better" and "general" in the original evaluation result, and after converting them into membership distribution vectors according to the above membership matrix, the membership degree of enterprise 1 under "better" is greater than that of enterprise 2 under "better", and the membership degree of enterprise 2 under "general" is greater than that of enterprise 1 under "general". In this way, the order of advantages and disadvantages in the original evaluation result is preserved, and the differentiation of the advantages and disadvantages of each enterprise is more obvious.

[0080] ​Finally, the method for calculating the objective function of the mathematical programming method is: according to each membership in the membership correspondence, the grade of the same qualitative index of each enterprise is evaluated, and the membership distribution vector of the same qualitative index of each enterprise belonging to different grades is converted respectively; the membership belonging to the same grade in the membership distribution vector of each enterprise is added, and the sum can represent the number of enterprises belonging to the same grade of the same qualitative index. If the membership under each grade is added to the most uniform, such as the standard deviation of the membership added under all grades reaches the minimum and remains stable, it is considered that the optimal goal has been reached. At this time, the membership matrix is used as the final membership matrix for subsequent conversion of the same qualitative index.

[0081] The embodiment sets the membership correspondence through the mathematical programming method to make the number of enterprises under each grade of the qualitative index relatively uniform, avoids the problem that the evaluation grades are clustered and cannot reflect the difference between enterprises and objective evaluation.

[0082] It should be noted that all data involved in the present application are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0083] Figure 4 A structural schematic diagram of an electronic device provided by the embodiment of the present application is shown in Figure 4 The device includes a processor 60, a memory 61, an input device 62 and an output device 63; the number of processors 60 in the device can be one or more, Figure 4 and an example of one processor 60 is taken; the processor 60, the memory 61, the input device 62 and the output device 63 in the device can be connected through a bus or other means, Figure 4 and an example of connection through a bus is taken.

[0084] The memory 61 is a kind of computer readable storage medium, which can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method in the embodiment of the present application. The processor 60 executes the software program, instruction and module stored in the memory 61, thereby performing various function applications and data processing of the device, that is, realizing the non-ferrous metal smelting enterprise pollution hidden danger multi-source heterogeneous index fusion evaluation method.

[0085] The memory 61 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, the memory 61 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 61 may further include memory remotely located relative to the processor 60, which can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0086] Input device 62 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the device. Output device 63 may include display devices such as a display screen.

[0087] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the multi-source heterogeneous index fusion assessment method for pollution risks in non-ferrous metal smelting enterprises according to any embodiment.

[0088] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0089] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0090] The program code embodied on the computer readable media can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0091] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0092] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, instead of limiting the present application; although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the technical solutions recorded in the above-mentioned embodiments can be modified or equivalent replacements can be made to some or all of the technical features; and the modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present application.

Claims

1. A method for fusion assessment of multi-source heterogeneous indicators of pollution risks in non-ferrous metal smelting enterprises, characterized in that, include: To obtain multiple assessment indicators for pollution risks in non-ferrous metal smelting enterprises, as well as the weight of each assessment indicator; Based on the evaluation level of each qualitative indicator among the multiple evaluation indicators, determine the membership degree distribution of each qualitative indicator under different levels; Based on the membership distribution and weight of each evaluation indicator, the comprehensive evaluation result of the enterprise is obtained.

2. The evaluation method according to claim 1, characterized in that, The step of determining the membership distribution of each qualitative indicator at different levels based on the assessed level of each qualitative indicator among the multiple evaluation indicators includes: For each qualitative indicator among multiple evaluation indicators: based on historical evaluation data and in accordance with the following principles, determine the membership degree correspondence for each level under the same qualitative indicator. The membership degree correspondence is used to express the membership degree of the same qualitative indicator at different levels when it is evaluated as various levels: Principle 1: When a certain qualitative indicator is evaluated to a certain level, the membership degree of the certain qualitative indicator under that certain level shall be assigned the highest membership degree, and the membership degree of the certain qualitative indicator under the adjacent levels of that certain level shall be assigned a non-zero membership degree. Principle 2: When a certain qualitative indicator is evaluated as a certain level, the sum of the membership degrees of the certain qualitative indicator under different levels is 1; Based on the membership degree correspondence of each qualitative indicator, the evaluation level of each qualitative indicator is converted into the membership degree distribution of each qualitative indicator under different levels.

3. The evaluation method according to claim 1, characterized in that, The step of determining the membership distribution of each qualitative indicator at different levels based on the assessed level of each qualitative indicator among the multiple evaluation indicators includes: When multiple enterprises cannot distinguish their level of superiority or inferiority through the same qualitative indicator among the multiple evaluation indicators, the membership degree correspondence of each level under the same qualitative indicator is determined so that the number of enterprises belonging to each level under the same qualitative indicator, as determined according to the membership degree correspondence, is evenly distributed. Among them, the membership degree correspondence is used to express the membership degree of the same qualitative index at different levels when it is evaluated as each level.

4. The evaluation method according to claim 3, characterized in that, The step of determining the membership degree correspondence of each level under the same qualitative indicator, so that the number of enterprises belonging to each level under the same qualitative indicator, as determined according to the membership degree correspondence, is evenly distributed, includes: Based on each membership degree in the membership degree correspondence, the level of the same qualitative indicator of each enterprise is evaluated and converted into the membership degree distribution vector of the same qualitative indicator of each enterprise under different levels. The membership degrees of enterprises belonging to the same level in the membership degree distribution vectors of each enterprise are added together, and the sum is used to represent the number of enterprises belonging to the same level for the same qualitative index.

5. The evaluation method according to claim 3, characterized in that, The step of determining the membership degree correspondence of each level under the same qualitative indicator, so that the number of enterprises belonging to each level under the same qualitative indicator, as determined according to the membership degree correspondence, is evenly distributed, includes: The membership degree correspondence of each level under the same qualitative index is determined by mathematical programming. In the mathematical programming method, when the same qualitative index is evaluated as a certain level, the certain level is taken as the central level. The variables to be planned are: the membership degree of the same qualitative index under each central level, and the membership degree under the adjacent levels of each central level. The constraints are as follows: the more enterprises that are evaluated to a certain level for the same qualitative indicator, the lower the membership degree of the same qualitative indicator under the certain level as the central level; the membership degree of the central level is greater than the membership degree of the adjacent level, the direction of change of the membership degree of the central level is opposite to the direction of change of the membership degree of the adjacent level, and the step size of the change of the central level is greater than the step size of the change of each adjacent level; the ranking of the superiority and inferiority of the levels evaluated by different enterprises is consistent with the order of superiority and inferiority represented by the membership degree of different enterprises under the same level. The planning objective is to redetermine the number of enterprises belonging to each level based on the new membership degree of each enterprise under different levels in each mathematical programming operation; the optimal objective is achieved when the number of enterprises belonging to each level is evenly distributed.

6. The evaluation method according to claim 1, characterized in that, Before obtaining the comprehensive evaluation result of the enterprise based on the membership distribution and weight of each evaluation indicator, the following steps are also included: Using the fuzzy evaluation method, the evaluation values ​​of each quantitative indicator among the multiple evaluation indicators are converted into the membership degree distribution of each quantitative indicator at different levels.

7. The evaluation method according to claim 1, characterized in that, The multiple assessment indicators include at least one of the following categories: personnel operation, equipment status, management system, and environmental characteristics.

8. The evaluation method according to claim 1, characterized in that, The weights of each evaluation indicator were determined using the analytic hierarchy process (AHP).

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the multi-source heterogeneous index fusion assessment method for pollution hazards in non-ferrous metal smelting enterprises as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the multi-source heterogeneous index fusion assessment method for pollution hazards in non-ferrous metal smelting enterprises as described in any one of claims 1-8.

Images