Multivariable intervention effect evaluation method and device in chlorination process titanium dioxide production process

Abstract

The embodiment of the invention provides a multivariable intervention effect evaluation method and device in a chlorination process titanium dioxide production process, and is applied to the technical field of artificial intelligence. The method comprises the following steps: acquiring intervention measures, confounding factors and result variables in the chlorination process titanium dioxide production process; on the basis of a tendency score matching method, according to each intervention measure, the confounding factor and the result variable, obtaining matched sample data; on the basis of a preset causal reasoning model, according to the matched sample data, respectively calculating an average intervention effect of each intervention measure on a result variable; wherein the estimator of the average intervention effect calculated by the preset causal reasoning model meets an estimator unbiased verification rule. In this way, quantitative evaluation can be carried out on the effects of different intervention measures, a scientific basis can be provided for decisions such as production process optimization, raw material selection, temperature and pressure control and the like, and the yield and quality of titanium dioxide can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, in particular to the field of artificial intelligence technology, and specifically to a method and device for evaluating the effect of multivariate intervention in a chloride process titanium dioxide production process. Background Art

[0002] With the rapid development of industrial production, the demand for titanium dioxide (TiO2) continues to increase. Outcome variables such as TiO2 yield and quality are influenced by multiple factors, including operational variables such as temperature, pressure, and raw material ratios. To optimize the production process and improve TiO2 yield and quality, multiple interventions are required. However, these interventions often interact with each other and are complexly associated with confounding factors such as equipment performance and operator variability.

[0003] Existing traditional causal inference methods, such as propensity score matching, can usually effectively evaluate the effect of a single intervention measure on the outcome variable. However, when faced with scenarios where multiple intervention measures act on multiple outcome variables simultaneously, traditional causal inference methods cannot accurately reveal the interactions and causal chains between multiple intervention measures, and cannot simultaneously consider the potential causal relationship and impact between multiple intervention measures and outcome variables. In addition, their ability to handle high-dimensional confounding factors is limited, which can easily lead to inaccurate estimation results and has certain limitations.

[0004] Therefore, how to effectively evaluate the causal effects between multiple intervention measures and multiple outcome variables has become an important issue in the optimization of titanium dioxide production process. Summary of the Invention

[0005] The present disclosure provides a method, device, equipment and storage medium for evaluating the effects of multivariate interventions in a chloride process titanium dioxide production process.

[0006] According to a first aspect of the present disclosure, a method for evaluating the effect of multivariate intervention in a chloride-process titanium dioxide production process is provided. The method comprises: To obtain the intervention measures, confounding factors and outcome variables in the chloride process of titanium dioxide production; Based on the propensity score matching method, matched sample data were obtained according to the intervention measures, confounding factors and outcome variables; Based on a preset causal inference model, the average intervention effect of each intervention measure on the outcome variable is calculated according to the matching sample data; wherein the estimated value of the average intervention effect calculated by the preset causal inference model meets the unbiasedness verification rule of the estimated value.

[0007] According to the above aspects and any possible implementation, an implementation is further provided, wherein the intervention measures, confounding factors, and outcome variables of the chloride process titanium dioxide production process are obtained, including: Obtain the various intervention measures in the production process of titanium dioxide by chloride method. The categorical variables of each intervention measure include intermediate intervention variables. Outcome intervention variables ; Obtaining confounding factors X in the chloride process titanium dioxide production process and performing normalization on the confounding factors; the confounding factors include equipment performance and operation differences; Obtain the result variable of the chloride process titanium dioxide production process, the categorical variable to which the result variable belongs includes the intermediate result variable and the final outcome variable ; The normalized confounding factors and the intermediate outcome variables are used to construct a logistic regression model, which includes: in, represents the intercept term, , , , Represent the corresponding confounding factors , , The coefficient of Represents intermediate result variables The coefficient of .

[0008] According to the above aspects and any possible implementation, an implementation is further provided, wherein obtaining matching sample data based on the propensity score matching method according to the intervention measures, confounding factors, and outcome variables includes: Based on the logistic regression model, the propensity score of each intervention measure is calculated using a preset propensity score algorithm; the preset propensity score algorithm includes: Among them, the propensity score Characterizes a given observation (confounding factor X, intermediate outcome variable ) Individuals accept the outcome intervention variable The conditional probability of According to the propensity score of each intervention measure, sample matching is performed on samples of different intervention groups and control group samples to obtain matched sample data; the intervention group samples are all samples that receive the intervention measures, and the control group samples are all samples that do not receive the intervention measures.

[0009] According to the above aspects and any possible implementation, a further implementation is provided, wherein the sample matching of different intervention group samples and control group samples is performed based on the propensity score of each intervention measure to obtain matched sample data, including: Based on the nearest neighbor matching algorithm, samples from different intervention groups were matched with those from the control group according to the propensity score of each intervention measure to obtain matched sample data.

[0010] According to the above aspects and any possible implementation, a further implementation is provided, wherein the preset causal inference model includes: in, represents a constant term, Represents the outcome intervention variable For the final outcome variable The direct effect of Represents intermediate result variables For the final outcome variable The indirect effect of Represents the final result variable The observed value of Represents the final result variable The estimated amount, Indicates the intermediate result variable The mapping function, represents inverse probability weighting based on the propensity score, It represents the average intervention effect of each intervention measure T in the matched sample data on the outcome variables of the intervention group samples and the control group samples in the chloride process of titanium dioxide production.

[0011] According to the above aspects and any possible implementation, an implementation is further provided, wherein the unbiasedness verification rule of the estimator includes: in, Represents the final result variable The observed value of Represents the final result variable The estimated amount.

[0012] According to the above aspects and any possible implementation, an implementation is further provided, wherein the intervention measures include temperature and pressure, and the result variables include the quality of the filtered material and the yield of titanium dioxide.

[0013] According to a second aspect of the present disclosure, a multivariate intervention effect evaluation device for a chloride process titanium dioxide production process is provided. The device comprises: The acquisition module is used to obtain the intervention measures, confounding factors and outcome variables in the chloride process of titanium dioxide production; A processing module, configured to obtain matched sample data based on the intervention measures, confounding factors, and outcome variables based on a propensity score matching method; A calculation module is used to calculate the average intervention effect of each intervention measure on the outcome variable based on the matching sample data based on a preset causal inference model; wherein the estimated amount of the average intervention effect calculated by the preset causal inference model meets the unbiasedness verification rule of the estimated amount.

[0014] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the above method when executing the program.

[0015] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0016] The embodiment of the present application provides a multivariate intervention effect evaluation method for the chloride titanium dioxide production process, which can obtain various intervention measures, confounding factors and outcome variables in the chloride titanium dioxide production process; then based on the propensity score matching method, obtain matching sample data according to each intervention measure, confounding factor and outcome variable; then based on a preset causal reasoning model, calculate the average intervention effect of each intervention measure on the outcome variable according to the matching sample data; wherein, the estimated amount of the average intervention effect calculated by the preset causal reasoning model meets the unbiasedness verification rule of the estimated amount; based on this, the above-mentioned multivariate intervention effect evaluation method can be widely used in the optimization and control of the titanium dioxide production process, especially in the case of multiple intervention measures and multiple outcome variables, by quantitatively evaluating the effects of different intervention measures, it can provide a scientific basis for decisions such as production process optimization, raw material selection, temperature and pressure control, and help improve the output and quality of titanium dioxide.

[0017] It should be understood that the contents described in the Summary of the Invention section are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are provided for a better understanding of the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which: Figure 1 A flow chart showing a multivariate intervention effect evaluation method for a chloride process titanium dioxide production process according to an embodiment of the present disclosure is shown; Figure 2 A schematic diagram of a simplified causal inference model of multivariate interventions under a potential outcome framework according to an embodiment of the present disclosure is shown; Figure 3 A schematic diagram of a multivariate intervention causal inference model in a titanium dioxide production scenario according to an embodiment of the present disclosure is shown; Figure 4 A schematic diagram showing simulation experiment results for verifying the unbiasedness of an estimator according to an embodiment of the present disclosure is shown; Figure 5 A block diagram of a multivariate intervention effect evaluation device for a chloride process titanium dioxide production process according to an embodiment of the present disclosure is shown; Figure 6 A block diagram of an exemplary electronic device capable of implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0019] To make the purpose, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.

[0020] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0021] In this disclosure, the effects of different intervention measures can be quantitatively evaluated to provide a scientific basis for decisions such as production process optimization, raw material selection, temperature and pressure control, and help improve the output and quality of titanium dioxide.

[0022] Figure 1 A flow chart of a multivariate intervention effect evaluation method 100 for a chloride-based titanium dioxide production process according to an embodiment of the present disclosure is shown.

[0023] In block 110 , various intervention measures, confounding factors, and outcome variables of the chloride process titanium dioxide production process are obtained.

[0024] In some embodiments, data collection and preprocessing are required to provide valid matching sample data for subsequent causal reasoning analysis. Specifically, various intervention measures, confounding factors, and outcome variables from the chloride process titanium dioxide production process are collected and preprocessed to provide valid matching sample data for subsequent causal reasoning analysis.

[0025] In some embodiments, the intervention measures, confounding factors, and outcome variables for obtaining the chloride process titanium dioxide production process specifically include: Obtain the various intervention measures in the production process of titanium dioxide by chloride method. The categorical variables of each intervention measure include intermediate intervention variables. and outcome intervention variables ; Obtain the confounding factor X in the chloride process titanium dioxide production process and perform normalization on the confounding factor; the confounding factor includes equipment performance and operation differences; Obtain the result variable of the chloride titanium dioxide production process. The categorical variables to which the result variable belongs include intermediate result variables. and the final outcome variable ; The normalized confounding factors and intermediate outcome variables are used to construct a logistic regression model, which includes: in, represents the intercept term, , , , Represent the corresponding confounding factors , , The coefficient of Represents intermediate result variables The coefficient of .

[0026] In some embodiments, the intervention measures include but are not limited to temperature and pressure, and the result variables include but are not limited to the amount of filtered material and the yield of titanium dioxide.

[0027] Specifically, data are collected from titanium dioxide production experiments or actual production systems. The data contain multiple confounding factors X (such as equipment performance, operator differences, etc.), intervention measures T (such as temperature, pressure, raw material ratio, etc.), and outcome variables Y (such as filtrate quality, titanium dioxide output, titanium dioxide quality, etc.).

[0028] like Figure 2 As shown, the categorical variables to which intervention T belongs can include intermediate intervention variables and outcome intervention variables , the categorical variables to which the outcome variable Y belongs can include intermediate outcome variables and the final outcome variable .

[0029] It should be noted that the intermediate intervention variable only intervenes in the intermediate outcome variable. In order not to confuse multiple intervention measures, the intermediate intervention variable and the outcome intervention variable are defined separately.

[0030] In some embodiments, the collected data may be preprocessed, for example, the confounding factor X may be normalized, that is, Normalization can ensure that the different dimensions of confounding factors have the same dimensions, which facilitates the subsequent logistic regression model building.

[0031] In some embodiments, a logistic regression model can be constructed based on the normalized confounding factor X, with the input variables being the confounding factor X and the intermediate outcome variable .

[0032] In some embodiments, the above-mentioned logistic regression model is applicable to the categorical variables of various intervention measures in the titanium dioxide production process, and the logistic regression model takes into account the influence of confounding factors such as equipment performance and operator differences in titanium dioxide production.

[0033] In block 120 , based on the propensity score matching method, matched sample data is obtained according to the intervention measures, confounding factors, and outcome variables.

[0034] In some embodiments, the propensity score matching method is used to perform logistic regression modeling on the relationship between intervention measures and confounding factors in the titanium dioxide production process; the propensity scores of multiple intervention measures are calculated based on the logistic regression model; and based on the propensity score of each intervention measure, samples of different intervention groups are matched with samples of the control group, wherein the intervention group is all samples that have received the intervention and the control group is all samples that have not received the intervention.

[0035] In some embodiments, the propensity score matching method is used to obtain matched sample data based on various intervention measures, confounding factors, and outcome variables, specifically including: Based on the logistic regression model, the propensity score of each intervention measure was calculated using a pre-defined propensity score algorithm. The pre-defined propensity score algorithm includes: Among them, the propensity score Characterizes a given observation (confounding factor X, intermediate outcome variable ) Individuals accept the outcome intervention variable The conditional probability of According to the propensity score of each intervention measure, sample matching was performed on samples of different intervention groups and control group samples to obtain matched sample data; the intervention group samples were all samples that received the intervention measures, and the control group samples were all samples that did not receive the intervention measures.

[0036] In some embodiments, the above-mentioned sample matching of different intervention group samples and control group samples based on the propensity score of each intervention measure to obtain matched sample data specifically includes: Based on the nearest neighbor matching algorithm, samples from different intervention groups were matched with those from the control group according to the propensity score of each intervention measure to obtain matched sample data.

[0037] In some embodiments, the propensity score of each intervention measure can be calculated based on the logistic regression model. Based on the logistic regression model, the intermediate outcome variable is introduced to calculate The correlation between each intervention and the confounding factors can be calculated by the propensity score It shows that the higher the score, the greater the correlation.

[0038] In some embodiments, sample matching can be performed between samples from different intervention groups and samples from the control group based on the propensity score of each intervention measure.

[0039] Specifically, based on the propensity score of each intervention measure, intervention measures (such as temperature, pressure) and The intervention group samples and the control group samples were matched with each other by nearest neighbor. For example, find all intervention group samples (i.e. ) and control group samples (i.e. ), using the nearest neighbor matching algorithm, each intervention group sample is paired with the control group sample with the closest propensity score. Sample matching ensures consistency in the distribution of confounding factors between the intervention group samples and the paired control group samples, providing effective matched sample data for subsequent causal inference analysis.

[0040] In box 130, based on the preset causal inference model and according to the matching sample data, the average intervention effect of each intervention measure on the outcome variable is calculated respectively; wherein the estimated value of the average intervention effect calculated by the preset causal inference model satisfies the unbiasedness verification rule of the estimated value.

[0041] In some embodiments, the categorical variables belonging to the intervention measures, such as temperature, pressure, raw material ratio, etc., can be combined with the categorical variables belonging to the outcome variables, such as the output and quality of titanium dioxide, etc., that is, the potential causal relationship between the intervention variables and the outcome variables can be combined to construct a causal reasoning model.

[0042] like Figure 3As shown in Figure 2, the relationship between intervention measures and confounding factors in the titanium dioxide production process is modeled, with each intervention measure in the titanium dioxide production process as input. is the input variable (intervention measure), which represents the acid hydrolysis temperature during the acid hydrolysis reaction; the acid hydrolyzed product after output is filtered through a series of processes to obtain , that is, the mass of the filtered material; when washing, is the input variable (intervention measure), which represents the water flow rate during the washing process, that is, the input variable Control the flow rate of washing water to affect the final output of titanium dioxide ,and right It also has an impact, so we need to consider the effect of each link in the production of titanium dioxide on the final titanium dioxide product.

[0043] In some embodiments, the preset causal inference model specifically includes: in, represents a constant term, Represents the outcome intervention variable For the final outcome variable The direct effect of Represents intermediate result variables For the final outcome variable The indirect effect of Represents the final result variable The observed value of Represents the final result variable The estimated amount, Indicates the intermediate result variable The mapping function, represents inverse probability weighting based on the propensity score, It represents the average intervention effect of each intervention measure T in the matched sample data on the outcome variables of the intervention group samples and the control group samples in the chloride process of titanium dioxide production.

[0044] In some embodiments, the above-mentioned unbiasedness verification rule of the estimator specifically includes: in, Represents the final result variable The observed value of Represents the final result variable The estimated amount.

[0045] In some embodiments, the total effect can be decomposed into direct effects and indirect effects transmitted through other dependent variables. In the case of mutual relationships between dependent variables, further decomposition of the total effect into direct effects and indirect effects transmitted through other dependent variables includes: considering right The potential impact of , construct a linear regression model: ; Through the regression coefficient and , respectively quantify the intervention right The direct effects and right The indirect effects of represents a constant term, express right The direct effect of express right indirect effects.

[0046] In some embodiments, the average treatment effect in the population (ATE) of each intervention measure T on the outcome variable Y in the titanium dioxide production process can be calculated.

[0047] In some embodiments, ATE represents the average difference in the outcome variable between the intervention group samples and the control group samples in all study subjects. Mathematically defined as: ATE=IE\left[{Y\left({T=1}\right)-Y\left({T=0}\right)}\right] ; Consider the impact of intermediate outcome variables on the final outcome variables, that is, right The impact of express right The impact of the intermediate intervening variables For the final outcome variable The impact of Intermediate intervening variables Only intervene in the intermediate outcome variables , the new adjustment result can be obtained as : ,in is the observed value of the final outcome variable, For the intermediate result variable The last term is the inverse probability weighting based on the propensity score. The estimated amount is: .

[0048] In some embodiments, when the influence between variables meets the linear condition, it is necessary to analyze whether the indirect effect of the decomposed variable on other variables is consistent with the estimate, that is, to perform an unbiased analysis of the estimate, and calculate the ATE of each intervention measure on the outcome variable (such as output, quality) separately.

[0049] In some embodiments, when the influence between variables satisfies a linear condition, analyzing whether the observed value of the decomposed variable on other variables is consistent with the estimated value includes: Estimates of the intervention effect on the adjusted outcome Is it consistent with the observation results Consistent, that is, verify that the following equation holds: , that is, only need to , we can prove that the estimator is unbiased.

[0050] Specifically, based on the above linear regression model: , it can be seen that and satisfies the linear relationship, and If it is not 0, right Existential Impact; Definition ,\Phi \left ( {{Y}_{1}} \right )=\frac {1} {{\beta}_{1}}\cdot \left [ {\Phi \left ( {{Y}_{2}} \right )-\Phi \left ( {X} \right )-\Phi \left ( {C} \right )} \right ] ; Substituting the above definition into , Obtain \(0 = \Phi(Y_1)\cdot\frac{T_2 - e(X_1,Y_1)}{e(X,Y_1)\cdot(1 - e(X,Y_1))}=\frac{1}{\beta_1}\cdot[\Phi(Y_2)-\Phi(X)-\Phi(C)]\cdot\frac{T_2 - e(X,Y_1)}{e(X,Y_1)\cdot(1 - e(X,Y_1))}\) ; If there exists , then: E\left ( {\frac {\frac {1} {{\beta}_{1}}\cdot \Phi \left ( {{Y}_{2}} \right )\cdot \left ( {{T}_{2}-e\left ( {X,{Y}_{1}} \right )} \right )} {e\left ( {X,{Y}_{1}} \right )\cdot \left ( {1-e\left ( {X,{Y}_{1}} \right )} \right )}| {X,{Y}_{1}} \right )} \right )}\cdot \left [ {E\left ( {{T}_{2}-e\left ( {X,{Y}_{1}} \right )|X,{Y}_{1}} \right )} \right ]=\frac {\frac {1} {{\beta}_{1}}\cdot \Phi \left ( {{Y}_{2}} \right )} {e\left ( {X,{Y}_{1}} \right )\left ( {1-e\left ( {X,{Y}_{1}} \right )} \right )}\cdot \left [ {E\left ( {{T}_{2}|X.{Y}_{1}} \right )-e\left ( {X,{Y}_{1}} \right )} \right ] ; Further get, ; It can be seen that the estimated results are consistent with the observed results and are unbiased.

[0051] like Figure 4 As shown in the figure, 20 groups of data can be randomly selected in the titanium dioxide production process for simulation experiments, and the ATE estimation results can be calculated, that is, the predicted average intervention effect, which is then compared with the actual average intervention effect. It can be seen that the ATE estimation results obtained by the multivariate intervention effect evaluation method for the chloride method titanium dioxide production process are close to the actual ATE.

[0052] In some embodiments, after verifying that the estimated value results are consistent with the observed value results and are unbiased, the causal relationship model and the matched sample data can be used, that is, the various intervention measures in the titanium dioxide production process can be used as the input of the causal relationship model to obtain the corresponding outcome variables, thereby calculating the ATE of each intervention measure on the outcome variable (such as output, quality) and evaluating the intervention effect of each intervention measure.

[0053] According to the embodiments of the present disclosure, the following technical effects are achieved: By obtaining the various intervention measures, confounding factors and outcome variables in the chloride process of titanium dioxide production; then based on the propensity score matching method, matching sample data are obtained according to the various intervention measures, confounding factors and outcome variables; then based on the preset causal inference model, the average intervention effect of each intervention measure on the outcome variable is calculated according to the matching sample data; among them, the estimate of the average intervention effect calculated by the preset causal inference model meets the unbiased verification rule of the estimate; based on this, the above-mentioned multivariate intervention effect evaluation method can be widely used in the optimization and control of the titanium dioxide production process, especially in the case of multiple intervention measures and multiple outcome variables. By quantitatively evaluating the effects of different intervention measures, it can provide a scientific basis for decisions such as production process optimization, raw material selection, temperature and pressure control, and help improve the output and quality of titanium dioxide.

[0054] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0055] The above is an introduction to the method embodiment. The following is a further explanation of the solution disclosed in the present disclosure through an apparatus embodiment.

[0056] Figure 5 FIG. 5 shows a block diagram of a multivariate intervention effect evaluation device 500 for a chloride process titanium dioxide production process according to an embodiment of the present disclosure. Figure 5 As shown, the device 500 includes: An acquisition module 510 is used to obtain various intervention measures, confounding factors, and outcome variables in the chloride process titanium dioxide production process; Processing module 520, for obtaining matched sample data based on the propensity score matching method according to the intervention measures, confounding factors and outcome variables; The calculation module 530 is used to calculate the average intervention effect of each intervention measure on the outcome variable based on the preset causal inference model and the matching sample data; wherein the estimated value of the average intervention effect calculated by the preset causal inference model meets the unbiasedness verification rule of the estimated value.

[0057] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0058] In the technical solutions disclosed herein, the acquisition, storage, and application of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0059] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0060] Figure 6 A block diagram of an exemplary electronic device 600 capable of implementing embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0061] The electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a ROM 602 or a computer program loaded from a storage unit 608 into a RAM 603. The RAM 603 may also store various programs and data required for the operation of the electronic device 600. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An I / O interface 605 is also connected to the bus 604.

[0062] Multiple components in the electronic device 600 are connected to the I / O interface 605, including an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0063] Computing unit 601 can be various general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. Computing unit 601 executes the various methods and processes described above, such as method 100. For example, in some embodiments, method 100 can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 608.

[0064] In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the method 100 described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the method 100 in any other appropriate manner (e.g., via firmware).

[0065] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0066] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0067] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0068] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0069] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0070] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0071] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0072] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A multivariate intervention effect evaluation method for a chloride process titanium dioxide production process, characterized in that: include: To obtain the intervention measures, confounding factors and outcome variables in the chloride process of titanium dioxide production; Based on the propensity score matching method, matched sample data were obtained according to the intervention measures, confounding factors and outcome variables; Based on a preset causal inference model, the average intervention effect of each intervention measure on the outcome variable is calculated according to the matching sample data; wherein the estimated value of the average intervention effect calculated by the preset causal inference model meets the unbiasedness verification rule of the estimated value.

2. The method according to claim 1, characterized in that The intervention measures, confounding factors and outcome variables for obtaining the chloride process titanium dioxide production process include: Obtain the various intervention measures in the production process of titanium dioxide by chloride method. The categorical variables of each intervention measure include intermediate intervention variables. and outcome intervention variables ; Obtaining confounding factors X in the chloride process titanium dioxide production process and performing normalization on the confounding factors; the confounding factors include equipment performance and operation differences; Obtain the result variable of the chloride process titanium dioxide production process, the categorical variable to which the result variable belongs includes the intermediate result variable and the final outcome variable ; The normalized confounding factors and the intermediate outcome variables are used to construct a logistic regression model, which includes: in, represents the intercept term, , , , Represent the corresponding confounding factors , , The coefficient of Represents intermediate result variables The coefficient of .

3. The method according to claim 2, characterized in that The propensity score matching method is based on the intervention measures, confounding factors and outcome variables to obtain the matched sample data, including: Based on the logistic regression model, the propensity score of each intervention measure is calculated using a preset propensity score algorithm; the preset propensity score algorithm includes: Among them, the propensity score Characterizes a given observation (confounding factor X, intermediate outcome variable ) Individuals accept the outcome intervention variable The conditional probability of According to the propensity score of each intervention measure, sample matching is performed on samples of different intervention groups and control group samples to obtain matched sample data; the intervention group samples are all samples that receive the intervention measures, and the control group samples are all samples that do not receive the intervention measures.

4. The method according to claim 3, characterized in that According to the propensity score of each intervention measure, samples of different intervention groups are matched with samples of control groups to obtain matched sample data including: Based on the nearest neighbor matching algorithm, samples from different intervention groups were matched with those from the control group according to the propensity score of each intervention measure to obtain matched sample data.

5. The method according to claim 3, characterized in that The preset causal reasoning model includes: in, represents a constant term, Represents the outcome intervention variable For the final outcome variable The direct effect of Represents intermediate result variables For the final outcome variable The indirect effect of Represents the final result variable The observed value of Represents the final result variable The estimated amount, Indicates the intermediate result variable The mapping function, represents inverse probability weighting based on the propensity score, It represents the average intervention effect of each intervention measure T in the matched sample data on the outcome variables of the intervention group samples and the control group samples in the chloride process of titanium dioxide production.

6. The method according to claim 5, characterized in that The unbiasedness verification rules of the estimator include: in, Represents the final result variable The observed value of Represents the final result variable The estimated amount.

7. The method according to any one of claims 2 to 6, characterized in that The interventions included temperature and pressure, and the outcome variables included filtrate mass and titanium dioxide production.

8. A multivariate intervention effect evaluation device for a chloride process titanium dioxide production process, characterized in that: include: The acquisition module is used to obtain the intervention measures, confounding factors and outcome variables in the chloride process of titanium dioxide production; A processing module, configured to obtain matched sample data based on the intervention measures, confounding factors, and outcome variables based on a propensity score matching method; A calculation module is used to calculate the average intervention effect of each intervention measure on the outcome variable based on the matching sample data based on a preset causal inference model; wherein the estimated amount of the average intervention effect calculated by the preset causal inference model meets the unbiasedness verification rule of the estimated amount.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.

Images