Industrial cluster early warning and optimization method, device, electronic equipment and storage medium

Abstract

The present disclosure provides an industrial cluster early warning and optimization method, device, electronic device and storage medium. The industrial cluster early warning and optimization method includes: based on the historical data of the industrial cluster, constructing a causal model including observed variables and latent variables, the latent variables including independent variables and dependent variables; calculating the latent variables of the current industrial cluster, and determining the independent variables to be optimized based on the early warning rules; based on the causal model, determining the observed variables that affect the independent variables to be optimized as the first observed variables; screening the second observed variables from the first observed variables, and adjusting the second observed variables. The technical solution of the embodiment of the present disclosure realizes early warning of industrial clusters through an early warning model. At the same time, by quantifying the latent variables, quantitatively analyzing the independent variables to be optimized and the corresponding observed variables, quantitative adjustment of industrial clusters is realized, and effective intervention and optimization of industrial clusters are completed.

Description

Background Art

[0002] Industrial clusters refer to a form of economic organization in which competitive enterprises in a certain industry and cooperative enterprises, suppliers, service providers, and related industrial manufacturers that interact with these enterprises gather in a specific region. The formation and development of industrial clusters help enterprises reduce costs, improve industrial efficiency and competitiveness, and promote regional economic development. The formation of industrial clusters is often based on factors such as resource endowment, market demand, and policy support, and is an important force in promoting industrial upgrading and economic development.

[0003] However, the development of industrial clusters faces many unstable factors, such as the instability of cooperative relationships, loose cooperation among enterprises, and the lack of a stable community of interests; factors such as market downturns and industry adjustments may lead to difficulties in production and operation of enterprises; technological risks, that is, industrial clusters may rely on certain technologies or skills, and once these technologies are surpassed, competitiveness may decline, etc.

[0004] In an industrial cluster, each member pursues his own interests. Competition among members will affect their own interests, but members seek cooperation for the sake of cooperation interests, which makes the decision-making of members on cooperation very complicated. The complexity of members' decision-making makes the performance of industrial clusters extremely unstable. In this context, how managers of industrial clusters, such as government departments and leading enterprises in the cluster, can warn of abnormal cluster performance and optimize performance to ensure the healthy development of industrial clusters has become a difficult problem in the development process of industrial clusters.

[0005] In the existing technology, there are few means to provide early warning of the performance of industrial clusters. In addition, most of the optimization methods for cluster performance provide some qualitative suggestions or correlation laws from the aspects of policy environment, scientific and technological innovation, brand building, etc., lacking early warning rules based on causal relationships and specific optimization methods, which makes it difficult to implement effective intervention in industrial clusters.

[0006] Therefore, how to achieve early warning of the performance of industrial clusters and optimize the performance of industrial clusters accordingly is an urgent problem to be solved in this field.

[0007] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0008] The purpose of the embodiments of the present disclosure is to provide an industrial cluster early warning and optimization method, device, electronic device and storage medium, thereby at least to a certain extent overcoming the problem that existing industrial cluster optimization methods lack causal-based early warning rules and specific optimization means, which makes it difficult to implement effective intervention in industrial clusters.

[0009] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0010] According to a first aspect of an embodiment of the present disclosure, an industrial cluster early warning and optimization method is provided, comprising: constructing a causal model including observed variables and latent variables based on historical data of industrial clusters, wherein the latent variables include independent variables and dependent variables; calculating the latent variables of the current industrial cluster, and determining the independent variables to be optimized based on early warning rules; based on the causal model, determining the observed variables that affect the independent variables to be optimized as the first observed variables; screening the second observed variables from the first observed variables, and adjusting the second observed variables.

[0011] In some example embodiments of the present disclosure, based on the aforementioned scheme, the calculation of the latent variables of the current industrial cluster and the determination of the independent variables to be optimized based on the early warning rules include: calculating the corresponding independent variables and dependent variables according to the current observed variables; calculating the value of each early warning rule according to the value of the independent variable; comparing the value of the early warning rule with the dependent variable to determine the early warning level corresponding to each rule; determining the number of votes for each variable according to the early warning level corresponding to each rule; and selecting the independent variable with the highest number of votes as the independent variable to be optimized.

[0012] In some example embodiments of the present disclosure, based on the aforementioned scheme, screening the second observed variable from the first observed variable includes: determining the urgency of adjusting the independent variable to be optimized based on the number of votes obtained by the independent variable to be optimized; determining the influence coefficient of each observed variable in the first observed variables on the independent variable to be optimized, and forming an observed variable sequence according to the influence coefficient from high to low; and randomly selecting an observed variable from a certain proportion of the observed variables before the observed variable sequence according to the adjustment urgency, as the second observed variable.

[0013] In some example embodiments of the present disclosure, based on the aforementioned scheme, adjusting the second observed variable includes: selecting the early warning rule including the independent variable to be optimized as the first early warning rule set, and the remaining early warning rules constitute the second early warning rule set; based on the value of the current independent variable, using the second early warning rule set to determine the performance value that the current industrial cluster should have; based on the performance value that the current industrial cluster should have, using the first early warning rule set to determine the adjustment target of the independent variable to be optimized; based on the adjustment target of the independent variable to be optimized, determine the target adjustment value of the second observed variable.

[0014] In some example embodiments of the present disclosure, based on the aforementioned scheme, the independent variables include capabilities of cluster members, differences among cluster members, and interactions within the cluster; and the dependent variables include cluster performance.

[0015] In some example embodiments of the present disclosure, based on the above scheme, the warning rule is:

[0016] Rule 1: y "# =m " x " +m # x # +C "# ;

[0017] Rule 2: y ") =m " x " +m ) x ) +C ") ;

[0018] Rule 3: y #) =m # x # +m ) x ) +C #) ;

[0019] Among them, x " 、x # 、x ) They represent the independent variables cluster member capabilities, cluster member differences, and interactions within clusters, respectively. " 、m # 、m ) They represent the linear fitting coefficients of the independent variables cluster member capabilities, cluster member differences, and interactions within clusters, respectively. "# ,y ") ,y #) is the cluster performance in rules 1, 2, and 3, C "# , C ") , C #) are the linear fitting constants in rules 1, 2 and 3 respectively.

[0020] In some example embodiments of the present disclosure, based on the above scheme, the calculating of the corresponding independent variables and dependent variables according to the current observed variables includes:

[0021] X=(k # +Δk # )y # +(k ) +Δk ) )y ) +(k5+Δk5)y5+…+(k7+Δk8)y7+e

[0022] Among them, X is the latent variable, y : is the value of the i-th observed variable, i = 1, 2, 3, ..., n, n is the number of observed variables corresponding to the latent variable X; k : for y :The corresponding regression coefficient; Δk : is the expected change in the impact of the corresponding ith observed variable, and e is the residual.

[0023] According to a second aspect of an embodiment of the present disclosure, an industrial cluster early warning and optimization device is provided, comprising: a causal model construction module, for constructing a causal model including observed variables and latent variables based on historical data of industrial clusters, wherein the latent variables include independent variables and dependent variables; a module for determining independent variables to be optimized, for calculating the latent variables of the current industrial cluster, and determining the independent variables to be optimized based on early warning rules; a first observed variable determination module, for determining, based on the causal model, observed variables that affect the independent variables to be optimized as first observed variables; and an adjustment module, for screening second observed variables from the first observed variables, and adjusting the second observed variables.

[0024] In some example embodiments of the present disclosure, based on the aforementioned scheme, the module for determining the independent variables to be optimized also includes: a latent variable calculation unit, used to calculate the corresponding independent variables and dependent variables according to the current observed variables; a calculation unit, used to calculate the value of each warning rule according to the value of the independent variable; a warning level determination unit, used to compare the value of the warning rule with the dependent variable to determine the warning level corresponding to each rule; a voting unit, used to determine the number of votes for each variable according to the warning level corresponding to each rule; and a selection unit, used to select the independent variable with the highest number of votes as the independent variable to be optimized.

[0025] In some example embodiments of the present disclosure, based on the aforementioned scheme, the adjustment module includes: an urgency determination unit, used to determine the urgency of adjusting the independent variable to be optimized according to the number of votes obtained by the independent variable to be optimized; an influence coefficient sorting unit, used to determine the influence coefficient of each observation variable in the first observation variables on the independent variable to be optimized, and form an observation variable sequence according to the influence coefficient from high to low; a second observation variable determination unit, used to randomly select an observation variable from a certain proportion of the observation variables before the observation variable sequence according to the adjustment urgency, as the second observation variable.

[0026] In some example embodiments of the present disclosure, based on the aforementioned scheme, the adjustment module includes: a classification unit, used to select early warning rules including the independent variables to be optimized as the first early warning rule set, and the remaining early warning rules constitute the second early warning rule set; a performance value determination unit, used to determine the performance value that the current industrial cluster should have based on the value of the current independent variable and using the second early warning rule set; a first adjustment target determination unit, used to determine the adjustment target of the independent variable to be optimized based on the performance value that the current industrial cluster should have and using the first early warning rule set; a second adjustment target determination unit, which determines the target adjustment value of the second observed variable based on the adjustment target of the independent variable to be optimized.

[0027] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the industrial cluster early warning and optimization method described in any one of the above is implemented.

[0028] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the industrial cluster early warning and optimization method according to any one of the above items is implemented.

[0029] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0030] The industrial cluster early warning and optimization method in the exemplary embodiment of the present disclosure, based on the historical data of the industrial cluster, constructs a causal model including observed variables and latent variables, the latent variables include independent variables and dependent variables; calculates the latent variables of the current industrial cluster, and determines the independent variables to be optimized based on the early warning rules; based on the causal model, determines the observed variables that affect the independent variables to be optimized as the first observed variables; selects the second observed variables from the first observed variables, and adjusts the second observed variables. The present disclosure proposes three factors that affect the performance of industrial clusters based on group intelligence and proposes a causal model of industrial clusters, combining "latent variables applicable to various industrial clusters" with "observed variables in specific industrial clusters", on the one hand, ensuring the operability and applicability of the model in different industrial clusters, and on the other hand, the scenario-based observed variables ensure the effectiveness and flexibility of the model across industrial clusters. Based on the constructed industrial cluster causal model, an early warning model for the performance of industrial clusters is proposed, and the early warning model includes early warning rules based on the causal model, voting mechanism, etc. Early warning of industrial clusters is achieved through the early warning model. At the same time, through the quantification of latent variables, quantitative analysis of the independent variables to be optimized and the corresponding observed variables, quantitative adjustment of industrial clusters can be achieved, and effective intervention and optimization of industrial clusters can be completed.

[0031] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present disclosure, and together with the specification, are used to explain the principles of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0033] Figure 1A schematic diagram schematically shows the process of the industrial cluster early warning and optimization method according to some embodiments of the present disclosure;

[0034] Figure 2 Schematically showing a schematic diagram of an industrial cluster causal model according to some embodiments of the present disclosure;

[0035] Figure 3 A schematic diagram schematically shows a process flow of a method for determining an independent variable to be optimized according to some embodiments of the present disclosure;

[0036] Figure 4 A schematic diagram schematically shows a flow chart of a method for determining a second observed variable according to some embodiments of the present disclosure;

[0037] Figure 5 A schematic diagram schematically shows a flow chart of a method for determining a second observed variable according to some embodiments of the present disclosure;

[0038] Figure 6 A schematic diagram schematically shows a flow chart of a method for calculating a second observed variable target adjustment value according to some embodiments of the present disclosure;

[0039] Figure 7 The schematic diagram shows the structure of the industrial cluster early warning and optimization device according to some embodiments of the present disclosure;

[0040] Figure 8 A schematic diagram schematically shows a structure of a computer system of an electronic device according to some embodiments of the present disclosure;

[0041] Fig. 9 A schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure is schematically shown.

[0042] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0043] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0044] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0045] Furthermore, the drawings are only schematic illustrations and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0046] In this example embodiment, firstly, an industrial cluster early warning and optimization method is provided, and the industrial cluster early warning and optimization can be applied to a server. Figure 1 The following is a schematic diagram of an industrial cluster early warning and optimization process according to some embodiments of the present disclosure. Figure 1 As shown, the industrial cluster early warning and optimization method may include the following steps:

[0047] Step S110: constructing a causal model including observed variables and latent variables based on the historical data of the industrial cluster, wherein the latent variables include independent variables and dependent variables;

[0048] Step S120, calculating the latent variables of the current industrial cluster, and determining the independent variables to be optimized based on the early warning rules;

[0049] Step S130: Based on the causal model, determine an observed variable that affects the independent variable to be optimized as a first observed variable;

[0050] Step S140: Filter a second observed variable from the first observed variable, and adjust the second observed variable.

[0051] Based on swarm intelligence, the present invention proposes three factors that affect the performance of industrial clusters and a causal model of industrial clusters, combining "latent variables applicable to various types of industrial clusters" with "observed variables in specific industrial clusters". On the one hand, the operability and applicability of the model in different industrial clusters are guaranteed, and on the other hand, the scenario-based observed variables ensure the effectiveness and flexibility of the model across industrial clusters. Based on the constructed causal model of industrial clusters, an early warning model for the performance of industrial clusters is proposed. The early warning model includes early warning rules and voting mechanisms based on causal models. Early warning of industrial clusters is achieved through the early warning model. At the same time, by quantifying latent variables, quantitatively analyzing the independent variables to be optimized and the corresponding observed variables, quantitative adjustment of industrial clusters is achieved, and effective intervention and optimization of industrial clusters are completed.

[0052] Next, the industrial cluster early warning and optimization method in this example embodiment will be further described.

[0053] Step S110: constructing a causal model including observed variables and latent variables based on the historical data of the industrial cluster, wherein the latent variables include independent variables and dependent variables;

[0054] Industrial clusters include many interrelated and interactive entities such as cooperative enterprises, suppliers, and service providers. In an exemplary embodiment of the present disclosure, in order to timely warn and optimize industrial clusters, firstly, a causal model of industrial clusters is constructed based on the intelligent emergence of the group, and according to the causal relationship in the causal model, abnormal industrial clusters are warned, and the influencing elements are determined. By optimizing the influencing elements, the overall optimization of the industrial cluster is achieved.

[0055] Specifically, the causal model described in the present disclosure includes two parts, one is latent variables and the other is observed variables. Observed variables are the way to observe latent variables. Latent variables include independent variables and dependent variables. There is a causal relationship between latent variables. In the model constructed by the present disclosure, latent variables are universal, while observed variables need to be determined in combination with specific scenarios.

[0056] In an exemplary embodiment of the present disclosure, latent variables include the capabilities of cluster members, differences among cluster members, interactions within clusters, and cluster performance. Specifically, they refer to:

[0057] Capabilities of cluster members: The comprehensive strength of members of an industrial cluster as competing enterprises or upstream and downstream enterprises, who continuously achieve their strategic goals, meet market demands and maintain competitive advantages by integrating internal resources and utilizing external opportunities.

[0058] Differences among cluster members: A combination of differences among members of an industrial cluster in terms of strategic objectives, business management, market environment, capabilities, human resources, etc. The relevant aspects of the specific differences need to be determined in combination with the industry to which the cluster belongs and the current situation.

[0059] Interaction within the cluster: interaction, competition and cooperation among members of the industrial cluster in terms of business, technology, information, resources, etc.

[0060] Cluster performance: Due to the close cooperation and common innovation atmosphere among enterprises within the cluster, the cluster has formed advantages and competitiveness in terms of cost, quality, service and innovation.

[0061] Among them, the ability of cluster members, the differences of cluster members, and the interactions within the cluster determine the performance of the cluster. Therefore, the ability of cluster members, the differences of cluster members, and the interactions within the cluster are independent variables in the latent variables, and the cluster performance is the dependent variable in the latent variables. Because the independent variables and dependent variables are proposed based on the process of collective intelligence emergence, they are not affected by specific industrial clusters and are general.

[0062] Although the latent variables in the causal model are universal, their corresponding observed variables need to be determined in combination with specific industrial clusters. By selecting observed variables in combination with industrial clusters, the independent variables and dependent variables can be accurately characterized and described.

[0063] The present disclosure uses multiple observed variables to jointly characterize a latent variable, which can make the characterization of the latent variable more comprehensive and improve the robustness of the framework. In addition, in the process of causal analysis of the framework, it is necessary to reasonably adjust the observed variables according to the analysis results to improve the accuracy of the selection of observed variables.

[0064] Specifically, in the process of analyzing industrial clusters, by collecting historical data of the corresponding industrial clusters and analyzing the historical data, we can determine the factors that affect the ability of cluster members, the differences among cluster members, the interactions within the clusters, and other latent variables within the industrial clusters, that is, determine the observed variables corresponding to each latent variable. Figure 2 As shown in the figure, the ellipse is the latent variable and the quadrilateral is the observed variable. By analyzing the historical data of industrial clusters, the observed variables used to observe the capabilities of latent variable cluster members include: member market share, employee size, employee satisfaction, hardware assets, intellectual property ownership, etc. The observed variables used to observe the differences among latent variable cluster members include: differences in the fields they belong to, differences in management, differences in market share, differences in the number of employees, etc. The observed variables used to observe the interactions within the latent variable cluster include: scale of cooperation, average number of meetings within the cluster, average scope of cooperation, degree of resource sharing, average number of business transactions, etc. The observed variables used to observe the performance of the latent variable cluster include: number of new member applications, total scale of cooperation, development speed, etc.

[0065] After the above-mentioned observed variables and latent variables are preliminarily determined, in order to more accurately characterize the causal relationship in the causal model, the present application collects the historical data of each observed variable in the above-mentioned industrial cluster causal model, supplements the relationship between the latent variables and sets the error variable to further improve the causal model, so as to analyze and verify the causal relationship and observed variables. The present disclosure does not limit the verification method of the causal model. For example, the structural equation model method can be used to verify the causal model, and the test process includes three parts: exploratory factor analysis, model testing, and confirmatory factor analysis. Specifically, first check whether the setting of the observed variables is reasonable (exploratory factor analysis); then check the matching degree between the relationship hypothesis and the data of the previous variables (model testing); finally, verify whether the observed variables of each latent variable are sufficient to characterize it, and whether there is sufficient discrimination between the latent variables (confirmatory factor analysis). If the test fails, it is necessary to correct the observed variables and start a new test again. Finally, the causal model that passes the test is a representation of the stable causal relationship behind the emergence of intelligence within the industrial cluster. The core purpose of the causal model is to obtain reasonable observed variables of latent variables in a specific industrial cluster and verify the causal relationship therein.

[0066] Step S120, calculating the latent variables of the current industrial cluster, and determining the independent variables to be optimized based on the early warning rules;

[0067] In order to provide early warning and optimization for industrial clusters, it is first necessary to quantify the latent variables, determine the status of the industrial clusters by monitoring the quantified latent variable values, and perform specific optimization based on the specific status of the industrial clusters.

[0068] Specifically, in the above causal model, the observed variables of each latent variable are screened by subjective experience and verified by objective tests, ensuring that each latent variable has a reasonable observation path. Here, according to the actual situation of the observed variables and industrial clusters, the measurement method of each latent variable in the industrial cluster can be clarified to provide support for the precision of causal relationships.

[0069] Calculating the corresponding independent variables and dependent variables based on the current observed variables includes:

[0070] X=(k # +Δk # )y # +(k ) +Δk ) )y ) +(k5+Δk5)y5+…+(k7+Δk8)y7+e

[0071] Among them, X is the latent variable, y :is the value of the i-th observed variable, i = 1, 2, 3, ..., n, n is the number of observed variables corresponding to the latent variable X; k : for y : The corresponding regression coefficient; Δk : is the expected change in the impact of the corresponding ith observed variable, and e is the residual.

[0072] In an exemplary embodiment of the present disclosure, in the tested causal model, it is assumed that the latent variable X is represented by y # ,y ) , y5 are jointly characterized by three observed variables, and the regression coefficients are k # , k ) , k5, the residual is negligible. The current situation of industrial clusters shows that the observed variable y ) The influence on the latent variable Y is expected to increase by Δk, then the quantitative formula of the latent variable X can be expressed as X = k # y # +(k ) +Δk)y ) +k5y5. The changes in the degree of influence or the adjustment of the observed variables in the industrial cluster need to be comprehensively determined in combination with the actual situation, and this disclosure does not limit this.

[0073] For a specific industrial cluster, the value of the observed variable can be directly determined. By determining the value of the observed variable, the corresponding cluster member's capabilities, the differences among cluster members, and the interactions within the cluster can be further calculated. With the quantitative formula for each independent variable, the value of the independent variable can be calculated based on the real-time data of the industrial cluster operation. On this basis, the causal relationship formula of the latent variable of the industrial cluster can be obtained.

[0074] Specifically, the independent variables of the causal relationship are: the ability of cluster members, the differences of cluster members, and the interactions within the cluster. The dependent variable is: cluster performance. For convenience, the independent variables cluster members’ ability, cluster members’ differences, and cluster interactions are represented by x " 、x # 、x ) The dependent variable cluster performance is represented by y, then the causal relationship of industrial clusters can be expressed as y = f(x " ,x # ,x ) ). The causal relationship can be obtained using various methods, which are not limited in the present disclosure. For example, the causal relationship can be determined using support vector machines, decision trees, neural networks, polynomial regression, etc.

[0075] The present invention obtains the values ​​of observed variables of industrial clusters at different times, calculates the values ​​of latent variables in the cluster, and determines the independent variables to be optimized by comparing the values ​​of latent variables with early warning rules, thereby achieving timely cluster optimization according to the status of the industrial cluster.

[0076] Specifically, the present disclosure adopts the method of clustering the capabilities x of the three independent variables into " , the difference of cluster members x # , the interaction within the cluster x ) In a pairing manner, a linear rule about the cluster performance y is formed. In the early warning process, the early warning rule needs to be accurate. The more variables are considered, the more the presented rule represents the general law of cluster operation, and the cluster performance loses its sensitivity to some variables or some situations. Therefore, the present disclosure adopts a pairing method of cluster member capabilities x0, cluster member differences x1, and cluster internal interactions x2. As for the fitting rules, since linear rules are worse than nonlinear fitting effects, more outliers will appear in actual use, so the early warning is more sensitive. The present disclosure improves the sensitivity of early warning by adopting linear rules.

[0077] Based on the historical data of cluster operation, a linear fit of two independent variables and cluster performance is performed to obtain the following early warning rules:

[0078] Rule 1: y 01 =m0x0+m1x1+C 01 .

[0079] Rule 2: y 02 =m0x0+m2x2+C 02 .

[0080] Rule 3: y 12 =m1x1+m2x2+C 12 .

[0081] Among them, x0, x1, and x2 represent the independent variables of cluster member capabilities, cluster member differences, and cluster internal interactions, respectively; m0, m1, and m2 represent the linear fitting coefficients of the independent variables of cluster member capabilities, cluster member differences, and cluster internal interactions, respectively; y 01 ,y 02 ,y 12 is the cluster performance in rules 1, 2, and 3, C 01 , C 02 , C 12 are the linear fitting constants in rules 1, 2 and 3 respectively.

[0082] Cluster performance at any time All warnings can be divided into four levels:

[0083] Level 1 warning: 0.5std<|Δ|≤std

[0084] Level 2 warning: std<|Δ|≤2std

[0085] Level 3 warning: 2std<|Δ|≤3std

[0086] Level 4 warning: 3std<|Δ|

[0087] Among them, the standard deviation std of cluster performance is calculated based on the historical data of industrial clusters.

[0088] Each of the three warning rules above votes based on the four-level warning. When |Δ|≤0.5std, no warning is issued. When Rule 1: y 01 =m0x0+m1x1+C 01 The occurrence of a warning means that the cluster has shown abnormal performance under the current cluster member's capabilities x0 and the cluster member's differences x1, which means that there is a problem with the independent variable cluster internal interaction x2, and vote for x2. Similarly, when a warning appears in rule 2, it means that the cluster has shown abnormal performance under the current cluster member's capabilities x0 and the cluster internal interaction x2, which means that there is a problem with the independent variable cluster member's differences x1, and vote for x1. When a warning appears in rule 3, it means that the cluster has shown abnormal performance under the current cluster member's differences x1 and the cluster internal interaction x2, which means that there is a problem with the current cluster member's capabilities x0, and vote for x0.

[0089] In addition, the warning level reflects the voting weight, and the first, second, third, and fourth warning levels correspond to voting weights of 1, 2, 3, and 4 respectively. The independent variable with the highest number of votes is the object to be optimized. If there is a tie, one of the independent variables with the highest number of votes is randomly selected for optimization.

[0090] For example, at time t, for cluster performance y t , assuming that the results calculated by rules 1, 2, and 3 correspond to the first-level warning, the second-level warning, and the first-level warning, respectively, then the number of votes corresponding to voting rule 1 is 1, and it votes for x2, and the number of votes x2 receives is 1. The number of votes corresponding to voting rule 2 is 2, and it votes for x1, and the number of votes x1 receives is 2. The number of votes corresponding to voting rule 3 is 1, and it votes for x0, and the number of votes x0 receives is 1. Therefore, according to the number of votes, x1 receives the most votes among the cluster members, and is the object that should be improved and optimized at present.

[0091] Specifically, refer to Figure 3 As shown, for the current industrial cluster, the method for determining the independent variable to be optimized may include the following steps S210 to S250:

[0092] S210, calculating corresponding independent variables and dependent variables according to current observed variables;

[0093] S220, calculating the value of each warning rule according to the value of the independent variable;

[0094] S230, comparing the value of the warning rule with the dependent variable to determine the warning level corresponding to each rule;

[0095] S240, determining the number of votes for each variable according to the warning level corresponding to each rule;

[0096] S250. Select the independent variable with the highest number of votes as the independent variable to be optimized.

[0097] Step S130: Based on the causal model, determine an observed variable that affects the independent variable to be optimized as a first observed variable;

[0098] Since observed variables are a way to observe latent variables, after determining the above-mentioned independent variables to be optimized, using the determined causal model to adjust the observed variables corresponding to the independent variables is an effective means of adjusting the above-mentioned variables to be optimized.

[0099] Specifically, assume that during the optimization process, X is the variable to be optimized. After calculating the quantitative expressions of each independent variable, for example, the quantitative formula of the independent variable X can be expressed as X=k1y1+(k2+Δk)y2+k3y3. The three observation variables y1, y2, and y3 are effective means to optimize X. That is, y1, y2, and y3 are the first observation variables.

[0100] Step S140: Filter a second observed variable from the first observed variable, and adjust the second observed variable.

[0101] For the first observed variable screened out, since the coefficients corresponding to the observed variables are different in the process of calculating the respective variables, that is, the values ​​of the observed variables have different effects on the independent variables, therefore, for the first observed variable, the observed variables with large influence coefficients are preferentially selected for adjustment to effectively optimize the value of the independent variable.

[0102] For the current independent variable to be optimized, X = k1y1+(k2+Δk)y2+k3y3+…+k n y n For example, different observed variables y1, y2, y3…y n The determinism of X can be expressed as Therefore, S is preferred. i In an exemplary embodiment of the present disclosure, directly select S iThe largest observed variable is used as the second observed variable.

[0103] Specifically, refer to Figure 4 As shown, the method for selecting the second observed variable from the first observed variable may include the following steps S310 to S320:

[0104] S310, determining the influence coefficient of each observed variable in the first observed variables on the independent variable to be optimized, and forming an observed variable sequence according to the influence coefficients from high to low;

[0105] S320. Select the first observed variable in the observed variable sequence as the second observed variable.

[0106] In another exemplary embodiment of the present disclosure, for the same independent variable to be optimized, if it obtains more votes in a vote, it means that its corresponding warning level is higher, the current industrial cluster needs to be optimized with higher urgency, and the urgency of adjusting the independent variable to be optimized is higher. For the independent variable to be optimized, the larger the influence coefficient of the adjusted observed variable is, the larger the range of the independent variable to be adjusted is. Therefore, the present disclosure further determines the observed variable to be adjusted as the second observed variable by combining the urgency of adjusting the independent variable.

[0107] Specifically, the number of votes an independent variable receives in a vote is V t , V t ≤4, considering the time dimension and depreciation coefficient d, if the variable to be optimized last time is the same as the variable to be optimized this time, it is considered that the number of votes V obtained by the latent variable this time is all =V t +dV t-1 , where V t-1 is the number of votes obtained in the last vote. d∈[0,1], the specific value is determined by the time sensitivity of the industry in which the specific industrial cluster is located.

[0108] For V all , the upper limit of the value range is 4(1+d), and the lower limit is 1. V all ∈[1,4(1+d)]. From the perspective of early warning, the most serious early warning latent variable receives 4(1+d) votes. For the latent variable to be optimized, calculate its V all for The urgency of this adjustment to the latent variable U∈[0,1].

[0109] According to the calculated urgency U, the second observation variable is determined, specifically:

[0110] When U ≥ 0.75, in S i Randomly select a y from the first 25% of the variablesi , with y i To optimize the latent variable, it is used as the second observed variable.

[0111] When U ≥ 0.5, in S i Randomly select a y from the first 50% of the variables i , with y i To optimize the latent variable, it is used as the second observed variable.

[0112] When U ≥ 0.25, in S i Randomly select a y from the first 75% of the variables i , with y i To optimize the latent variable, it is used as the second observed variable.

[0113] Otherwise, randomly select a i , with y i To optimize the latent variable, it is used as the second observed variable.

[0114] It is worth noting that the above urgency values ​​are related to S i The application does not limit the selection of the numerical range, as long as the value of the urgency is larger, the selected S i The smaller the numerical range, the more it falls within the protection scope of this application.

[0115] Specifically, refer to Figure 5 As shown, the method for selecting the second observed variable from the first observed variable may include the following steps S410 to S430:

[0116] S410, determining the urgency of adjusting the independent variable to be optimized according to the number of votes obtained by the independent variable to be optimized;

[0117] S420, determining the influence coefficient of each observed variable in the first observed variables on the independent variable to be optimized, and forming an observed variable sequence according to the influence coefficients from high to low;

[0118] S430. According to the adjustment urgency, randomly select an observation variable from a certain proportion of observation variables before the observation variable sequence as the second observation variable.

[0119] Select the observed variable y i As the second observed variable, it is necessary to adjust the second observed variable to optimize the industrial cluster. However, it is difficult to determine the degree of intervention and optimization. For example, the observed variable y i The present disclosure proposes a method for determining the degree of adjustment of the second observed variable. Specifically, the present disclosure selects the observed variable y i Then, according to the above warning rules, determine y idegree of adjustment.

[0120] Specifically, refer to Figure 6 As shown, for the second observed variable, the method for calculating its target adjustment value may include the following steps S510 to S540:

[0121] S510, selecting the warning rule including the independent variable to be optimized as the first warning rule set, and the remaining warning rules constitute the second warning rule set;

[0122] The present disclosure pairs the three independent variables, the ability x0 of cluster members, the difference x1 of cluster members, and the interaction x2 within the cluster, to form three different warning rules. Therefore, for any independent variable (X∈{x0,x1,x2}), there must be a rule involving the independent variable X and a warning rule not involving the independent variable X. For example, when X=x0, the above rule 3 is a warning rule not involving X, and rules 1 and 2 are warning rules involving X.

[0123] Because, the present disclosure first classifies all warning rules based on whether they include the independent variables to be optimized based on the determined independent variables to be optimized. The warning rules including the independent variables to be optimized are selected as the first warning rule set, and the remaining warning rules form the second warning rule set. For an industrial cluster including three independent variables, a first warning rule set including two warning rules and a second warning rule set including one warning rule are inevitably formed.

[0124] S520, based on the value of the current independent variable, using the second early warning rule set to determine the performance value that the current industrial cluster should have;

[0125] As described above, the second warning rule set is composed of warning rules that do not include the independent variable to be optimized. In order to optimize the independent variable to be optimized, the present disclosure uses the second warning rule set that does not involve the independent variable to be optimized X as a benchmark, and obtains the performance Y that the current industrial cluster should have according to the current values ​​of the remaining independent variables.

[0126] For example, for the three independent variables of cluster member capabilities x0, cluster member differences x1, and cluster internal interactions x2, the independent variable to be optimized X = x0, the current independent variables cluster member differences x1 and cluster internal interactions x2 are brought into early warning rule 3: y 12 =m1x1+m2x2+C 12 , and the obtained y 12 Performance Y that the current industrial cluster should have.

[0127] S530, based on the performance value that the current industrial cluster should have, using the first early warning rule set to determine the adjustment target of the independent variable to be optimized;

[0128] For the three independent variables, the ability of cluster members x0, the difference of cluster members x1, and the interaction within the cluster x2, the first warning rule set includes two warning rules. For the two warning rules involving X, by bringing in the performance Y that the current industrial cluster should have and the current values ​​of other non-X independent variables, the value that the latent variable X should present can be obtained. and but is the adjustment target of the independent variable X.

[0129] S540: Determine a target adjustment value of a second observed variable based on the adjustment target of the independent variable to be optimized.

[0130] Since there is a direct quantitative relationship between the independent variable and the corresponding observed variable, After the adjustment target of the independent variable X is obtained, the target adjustment value of the second observed variable can be determined based on the numerical relationship between the independent variable X and the second observed variable.

[0131] Specifically, for example, the independent variable X and the observed variable y i The relationship between them can be expressed as X = k1y1 + k2y2 + k3y3 + ... + k n y n , for the observed variable y i , which can be expressed as:

[0132]

[0133] Therefore, when the adjustment target of the independent variable X to be optimized is calculated Then, for any second observed variable y i , all other observed variables y j (j≠i) keeps the current value unchanged, and the target adjustment value of the second observed variable is calculated by the above formula as the adjustment target of the second observed variable at the next moment.

[0134] The present disclosure performs continuous dynamic monitoring and early warning optimization on the industrial cluster according to the status of the industrial cluster, and cyclically executes corresponding early warning and optimization steps S120-S140 according to the status of the industrial cluster.

[0135] In the actual industrial cluster optimization process, the industrial cluster can be optimized regularly. In an exemplary embodiment of the present disclosure, assuming that the cluster optimization step is τ, then for time t, if t|τ (t divides τ), then the industrial cluster needs to be optimized at time t, and the corresponding optimization is performed based on the above industrial cluster optimization measures, otherwise, the corresponding optimization does not need to be performed at the current time.

[0136] In summary, based on the historical data of industrial clusters, a causal model including observed variables and latent variables is constructed, wherein the latent variables include independent variables and dependent variables; the latent variables of the current industrial cluster are calculated, and the independent variables to be optimized are determined based on the early warning rules; based on the causal model, the observed variables affecting the independent variables to be optimized are determined as the first observed variables; the second observed variables are screened from the first observed variables, and the second observed variables are adjusted. The present disclosure proposes three factors that affect the performance of industrial clusters based on group intelligence and proposes a causal model of industrial clusters, combining "latent variables applicable to various industrial clusters" with "observed variables in specific industrial clusters". On the one hand, the operability and applicability of the model in different industrial clusters are guaranteed, and on the other hand, the scenario-based observed variables ensure the effectiveness and flexibility of the model across industrial clusters. Based on the constructed causal model of industrial clusters, an early warning model for the performance of industrial clusters is proposed, and the early warning model includes early warning rules based on the causal model, voting mechanism, etc. Early warning of industrial clusters is realized through the early warning model. At the same time, by quantifying the latent variables, quantitatively analyzing the independent variables to be optimized and the corresponding observed variables, quantitative adjustment of industrial clusters is realized, and effective intervention and optimization of industrial clusters are completed.

[0137] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0138] In addition, in this exemplary embodiment, an industrial cluster early warning and optimization device is also provided. Figure 7 As shown, the industrial cluster early warning and optimization device includes: a causal model construction module, which is used to construct a causal model including observed variables and latent variables based on historical data of industrial clusters, and the latent variables include independent variables and dependent variables; a module for determining independent variables to be optimized, which is used to calculate the latent variables of the current industrial cluster and determine the independent variables to be optimized based on early warning rules; a first observed variable determination module, which is used to determine the observed variables that affect the independent variables to be optimized based on the causal model as the first observed variables; and an adjustment module, which is used to screen the second observed variables from the first observed variables and adjust the second observed variables.

[0139] In some example embodiments of the present disclosure, based on the aforementioned scheme, the module for determining the independent variables to be optimized also includes: a latent variable calculation unit, used to calculate the corresponding independent variables and dependent variables according to the current observed variables; a calculation unit, used to calculate the value of each warning rule according to the value of the independent variable; a warning level determination unit, used to compare the value of the warning rule with the dependent variable to determine the warning level corresponding to each rule; a voting unit, used to determine the number of votes for each variable according to the warning level corresponding to each rule; and a selection unit, used to select the independent variable with the highest number of votes as the independent variable to be optimized.

[0140] In some example embodiments of the present disclosure, based on the aforementioned scheme, the adjustment module includes: an urgency determination unit, used to determine the urgency of adjusting the independent variable to be optimized according to the number of votes obtained by the independent variable to be optimized; an influence coefficient sorting unit, used to determine the influence coefficient of each observation variable in the first observation variables on the independent variable to be optimized, and form an observation variable sequence according to the influence coefficient from high to low; a second observation variable determination unit, used to randomly select an observation variable from a certain proportion of the observation variables before the observation variable sequence according to the adjustment urgency, as the second observation variable.

[0141] In some example embodiments of the present disclosure, based on the above solution, the adjustment module includes:

[0142] The influence coefficient sorting unit is used to determine the influence coefficient of each observed variable in the first observed variables on the independent variable to be optimized, and form an observed variable sequence from high to low according to the influence coefficient; the second observed variable determination unit selects the observed variable at the first position in the observed variable sequence as the second observed variable.

[0143] In some example embodiments of the present disclosure, based on the aforementioned scheme, the adjustment module includes: a classification unit, used to select early warning rules including the independent variables to be optimized as the first early warning rule set, and the remaining early warning rules constitute the second early warning rule set; a performance value determination unit, used to determine the performance value that the current industrial cluster should have based on the value of the current independent variable and using the second early warning rule set; a first adjustment target determination unit, used to determine the adjustment target of the independent variable to be optimized based on the performance value that the current industrial cluster should have and using the first early warning rule set; a second adjustment target determination unit, which determines the target adjustment value of the second observed variable based on the adjustment target of the independent variable to be optimized.

[0144] The specific details of each module and its sub-units of the above-mentioned industrial cluster early warning and optimization device have been described in detail in the corresponding industrial cluster early warning and optimization method, so they will not be repeated here.

[0145] It should be noted that although several modules or units of the industrial cluster early warning and optimization device are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0146] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above-mentioned industrial cluster early warning and optimization method is also provided.

[0147] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware embodiments, complete software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0148] Refer to the following Figure 8 An electronic device according to such an embodiment of the present disclosure is described. Figure 8 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0149] like Figure 8 As shown, the electronic device 400 is in the form of a general computing device. The components of the electronic device 400 may include but are not limited to: the at least one processing unit 410, the at least one storage unit 420, a bus 430 connecting different system components (including the storage unit 420 and the processing unit 410), and a display unit 440.

[0150] The storage unit stores a program code, which can be executed by the processing unit 810, so that the processing unit 810 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit can perform the following steps: Figure 1 Step S110 shown in the figure, based on the historical data of the industrial cluster, construct a causal model including observed variables and latent variables, the latent variables include independent variables and dependent variables; step S120, calculate the latent variables of the current industrial cluster, and determine the independent variables to be optimized based on the early warning rules; step S130, based on the causal model, determine the observed variables that affect the independent variables to be optimized as the first observed variables; step S140, select the second observed variables from the first observed variables, and adjust the second observed variables.

[0151] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 421 and / or a cache memory unit 422 , and may further include a read-only memory unit (ROM) 423 .

[0152] The storage unit 420 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0153] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0154] The electronic device 400 may also communicate with one or more external devices 470 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 400, and / or communicate with any device that enables the electronic device 400 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 450. In addition, the electronic device 400 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 460. As shown, the network adapter 460 communicates with other modules of the electronic device 400 via a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0155] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0156] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible embodiments, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to perform the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0157] refer to Fig. 9 As shown, a program product for implementing the above-mentioned contribution calculation method based on privacy computing according to an embodiment of the present disclosure is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited to this. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, apparatus, or device.

[0158] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0159] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0160] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0161] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0162] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0163] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0164] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0165] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An industrial cluster early warning and optimization method, characterized in that: include: Step S110: constructing a causal model including observed variables and latent variables based on the historical data of the industrial cluster, wherein the latent variables include independent variables and dependent variables; Step S120, calculating the latent variables of the current industrial cluster, and determining the independent variables to be optimized based on the early warning rules; Step S130: Based on the causal model, determine an observed variable that affects the independent variable to be optimized as a first observed variable; Step S140, selecting a second observed variable from the first observed variable, and adjusting the second observed variable; The calculation of the latent variables of the current industrial cluster and the determination of the independent variables to be optimized based on the early warning rules include: S210, calculating corresponding independent variables and dependent variables according to current observed variables; S220, calculating the value of each warning rule according to the value of the independent variable; S230, comparing the value of the warning rule with the dependent variable to determine the warning level corresponding to each rule; S240, determining the number of votes for each variable according to the warning level corresponding to each rule; S250, selecting the independent variable with the highest number of votes as the independent variable to be optimized; Screening the second observed variable from the first observed variable comprises: S410, determining the urgency of adjusting the independent variable to be optimized according to the number of votes obtained by the independent variable to be optimized; S420, determining the influence coefficient of each observed variable in the first observed variables on the independent variable to be optimized, and forming an observed variable sequence according to the influence coefficients from high to low; S430, randomly selecting an observation variable from a certain proportion of observation variables before the observation variable sequence according to the adjustment urgency, as the second observation variable; Adjusting the second observed variable includes: S510, selecting the warning rule including the independent variable to be optimized as the first warning rule set, and the remaining warning rules constitute the second warning rule set; S520, based on the value of the current independent variable, using the second early warning rule set to determine the performance value that the current industrial cluster should have; S530, based on the performance value that the current industrial cluster should have, using the first early warning rule set to determine the adjustment target of the independent variable to be optimized; S540: Determine a target adjustment value of a second observed variable based on the adjustment target of the independent variable to be optimized.

2. The early warning and optimization method according to claim 1, characterized in that: The independent variables include the capabilities of cluster members, the differences among cluster members, and the interactions within the cluster; the dependent variables include cluster performance.

3. The early warning and optimization method according to claim 2, characterized in that: The warning rules are: Rule 1: y 01 =m0x0+m1x1+C 01 ; Rule 2: y 02 =m0x0+m2x2+C 02 ; Rule 3: y 12 =m1x1+m2x2+C 12 ; Among them, x0, x1, and x2 represent the independent variables of cluster member capabilities, cluster member differences, and cluster internal interactions, respectively; m0, m1, and m2 represent the linear fitting coefficients of the independent variables of cluster member capabilities, cluster member differences, and cluster internal interactions, respectively; y 01 ,y 02 ,y 12 is the cluster performance in rules 1, 2, and 3, C 01 , C 02 , C 12 are the linear fitting constants in rules 1, 2 and 3 respectively.

4. The early warning and optimization method according to claim 1, characterized in that: The calculation of the corresponding independent variables and dependent variables according to the current observed variables includes: X=(k1+Δk1)y1+(k2+Δk2)y2+(k3+Δk3)y3+…+(k n +Δk n )y n +e Among them, X is the latent variable, y i is the value of the i-th observed variable, i = 1, 2, 3, ..., n, n is the number of observed variables corresponding to the latent variable X; k i for y i The corresponding regression coefficient; Δk i is the expected change in the impact of the corresponding ith observed variable, and e is the residual.

5. An industrial cluster early warning and optimization device, characterized in that: include: A causal model building module, used to build a causal model including observed variables and latent variables based on the historical data of the industrial cluster, wherein the latent variables include independent variables and dependent variables; The module for determining the independent variables to be optimized is used to calculate the latent variables of the current industrial cluster and determine the independent variables to be optimized based on the early warning rules; A first observed variable determination module, used to determine, based on the causal model, an observed variable that affects the independent variable to be optimized as a first observed variable; an adjustment module, configured to select a second observed variable from the first observed variable and adjust the second observed variable; The module for determining the independent variable to be optimized also includes: The latent variable calculation unit is used to calculate the corresponding independent variables and dependent variables according to the current observed variables; A calculation unit, used to calculate the value of each warning rule according to the value of the independent variable; An early warning level determination unit, used to compare the value of the early warning rule with the dependent variable to determine the early warning level corresponding to each rule; The voting unit is used to determine the number of votes for each variable according to the warning level corresponding to each rule; A selection unit, used to select the independent variable with the highest number of votes as the independent variable to be optimized; The adjustment module comprises: An urgency determination unit, configured to determine the urgency of adjusting the independent variable to be optimized according to the number of votes obtained by the independent variable to be optimized; An influence coefficient sorting unit, used to determine the influence coefficient of each observed variable in the first observed variables on the independent variable to be optimized, and form an observed variable sequence according to the influence coefficient from high to low; A second observed variable determination unit is used to randomly select an observed variable from a certain proportion of observed variables before the observed variable sequence according to the adjustment urgency, as the second observed variable; The adjustment module comprises: A classification unit, used for selecting the warning rules including the independent variables to be optimized as the first warning rule set, and the remaining warning rules constitute the second warning rule set; a performance value determination unit, for determining the performance value that the current industrial cluster should have based on the value of the current independent variable and using the second early warning rule set; A first adjustment target determination unit, used to determine the adjustment target of the independent variable to be optimized based on the performance value that the current industrial cluster should have and using the first early warning rule set; The second adjustment target determination unit determines a target adjustment value of the second observed variable based on the adjustment target of the independent variable to be optimized.

6. An electronic device comprising: processor; as well as A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the industrial cluster early warning and optimization method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the industrial cluster early warning and optimization method according to any one of claims 1 to 4.

Images