Method, device, storage medium and processor for calculating the concentration of raffinate hydrogen peroxide

Abstract

The present invention provides a method, device, storage medium and processor for calculating the concentration of the residual hydrogen peroxide, which belongs to the field of chemical technology. The calculation method includes: detecting the current data of other sites other than the residual hydrogen peroxide concentration of the hydrogen peroxide device; calculating the residual hydrogen peroxide concentration of the hydrogen peroxide device based on the current data of the other sites and the hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by neural network training of the fluctuation historical data of the residual hydrogen peroxide concentration and the historical data of other sites related to the fluctuation historical data of the residual hydrogen peroxide concentration. The present invention does not rely on manpower and provides guarantee for safe operation.

Description

Technical Field

[0001] The present invention relates to the field of chemical technology, and in particular to a method, device, storage medium and processor for calculating the concentration of raffinate hydrogen peroxide. Background Art

[0002] Although the anthraquinone method of hydrogen peroxide production has the characteristics of small scale and low structural complexity compared with other chemical processes, due to its poor reaction stability and high risk of process products, accidents often occur in the process equipment, causing serious losses of personnel and property. In the anthraquinone method of hydrogen peroxide process, the hydrogen peroxide content in the raffinate is a key indicator for the safe operation of the device. If the concentration of hydrogen peroxide is too high, it will decompose in the hydrogenation tower to produce oxygen, which will be mixed with hydrogen and cause gas phase explosion. The control of the concentration of raffinate hydrogen peroxide is of great significance to safe production. There is no online measurement equipment for the concentration of raffinate hydrogen peroxide in the existing hydrogen peroxide extraction tower, and manual sampling is used. This method has a long sampling interval and high cost. In addition, the automatic hydrogen peroxide concentration tester of the raffinate in the existing anthraquinone method of hydrogen peroxide production process is relatively expensive, has a large measurement error, and the emissions cannot meet environmental protection standards, which does not meet the actual needs of my country's hydrogen peroxide enterprises. Summary of the invention

[0003] The object of the present invention is to provide a method, device, storage medium and processor for calculating the concentration of raffinate hydrogen peroxide, which do not rely on manpower and provide guarantee for safe operation.

[0004] In order to achieve the above object, the present invention provides a method for calculating the raffinate hydrogen peroxide concentration of a hydrogen peroxide device, the calculation method comprising: detecting current data of other sites other than the raffinate hydrogen peroxide concentration of the hydrogen peroxide device; calculating the raffinate hydrogen peroxide concentration of the hydrogen peroxide device according to the current data of the other sites and based on a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by performing neural network training on the fluctuation historical data of the raffinate hydrogen peroxide concentration and the historical data of other sites related to the fluctuation historical data of the raffinate hydrogen peroxide concentration.

[0005] Preferably, the fluctuation historical data is historical data in which the duration of abnormal operation data within a preset time period is greater than a preset percentage of the preset time period.

[0006] Preferably, the hydrogen peroxide concentration prediction model is obtained by: acquiring the fluctuation history data of the raffinate hydrogen peroxide concentration; calculating the grayscale correlation coefficient between the fluctuation history data and the history data of each site at other sites at the same operating time; extracting the fluctuation history data and the history data of other sites whose grayscale correlation coefficient is greater than a correlation coefficient threshold value related to the site type; and training based on the extracted data and a neural network to obtain the hydrogen peroxide concentration prediction model.

[0007] Preferably, the grayscale correlation coefficient is calculated by the following formula:

[0008] Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

[0009] Preferably, other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold related to the site type include: the gas phase temperature after the oxidation exhaust heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation exhaust heat exchanger, and the opening of the extraction tower top boundary position control regulating valve.

[0010] Preferably, the neural network training uses MATLAB neural network.

[0011] Preferably, 70% of the historical data of fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data of fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

[0012] The present invention also provides a device for calculating the raffinate hydrogen peroxide concentration of a hydrogen peroxide device, the calculation device comprising: a detection unit and a processing unit, wherein the detection unit is used to detect current data of other sites other than the raffinate hydrogen peroxide concentration of the hydrogen peroxide device; the processing unit is used to calculate the raffinate hydrogen peroxide concentration of the hydrogen peroxide device based on the current data of the other sites and a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by performing neural network training on the fluctuation historical data of the raffinate hydrogen peroxide concentration and the historical data of other sites related to the fluctuation historical data of the raffinate hydrogen peroxide concentration.

[0013] Preferably, the fluctuation historical data is historical data in which the duration of abnormal operation data within a preset time period is greater than a preset percentage of the preset time period.

[0014] Preferably, the hydrogen peroxide concentration prediction model is obtained by: acquiring the fluctuation history data of the raffinate hydrogen peroxide concentration; calculating the grayscale correlation coefficient between the fluctuation history data and the history data of each site at other sites at the same operating time; extracting the fluctuation history data and the history data of other sites whose grayscale correlation coefficient is greater than a correlation coefficient threshold value related to the site type; and training based on the extracted data and a neural network to obtain the hydrogen peroxide concentration prediction model.

[0015] Preferably, the grayscale correlation coefficient is calculated by the following formula:

[0016] Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

[0017] Preferably, other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold related to the site type include: the gas phase temperature after the oxidation exhaust heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation exhaust heat exchanger, and the opening of the extraction tower top boundary position control regulating valve.

[0018] Preferably, the neural network training uses MATLAB neural network.

[0019] Preferably, 70% of the historical data of fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data of fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

[0020] The present invention also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device.

[0021] The present invention also provides a processor for running a program, wherein the program is used to execute the above-mentioned method for calculating the concentration of raffinate hydrogen peroxide in the hydrogen peroxide device when it is run.

[0022] Through the above technical scheme, the method, device, storage medium and processor for calculating the concentration of raffinate hydrogen peroxide provided by the present invention are adopted, and the calculation method includes: detecting the current data of other sites other than the concentration of raffinate hydrogen peroxide of the hydrogen peroxide device; according to the current data of the other sites, based on the hydrogen peroxide concentration prediction model, the concentration of raffinate hydrogen peroxide of the hydrogen peroxide device is calculated, wherein the hydrogen peroxide concentration prediction model is a model obtained by training a neural network with the fluctuation historical data of the concentration of raffinate hydrogen peroxide and the historical data of other sites related to the fluctuation historical data of the concentration of raffinate hydrogen peroxide. The present invention does not rely on manpower and provides guarantee for safe operation.

[0023] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0025] Figure 1 is a flow chart of a method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device provided in one embodiment of the present invention;

[0026] Figure 2 is a flow chart of a method for establishing a hydrogen peroxide concentration prediction model provided by one embodiment of the present invention;

[0027] Figure 3 It is a structural block diagram of a device for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device provided by one embodiment of the present invention.

[0028] Description of Reference Numerals

[0029] 1 Detection unit 2 Processing unit DETAILED DESCRIPTION

[0030] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0031] Embodiment 1

[0032] Figure 1 1 is a flow chart of a method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to an embodiment of the present invention. Figure 1 As shown, the calculation method includes:

[0033] Step S11, detecting current data of other points other than the concentration of the residual hydrogen peroxide of the hydrogen peroxide device;

[0034] Specifically, the sites may be different measurement objects of different chemical plants, such as the oxygen content of hydrogenation tower tail gas, the oxygen content of oxidation tower tail gas, or the hydrogen peroxide concentration of raffinate, etc., or different measurement objects in the same chemical plant, for example, for a hydrogen peroxide plant, especially anthraquinone hydrogen peroxide plant, the sites may be the hydrogen peroxide concentration of raffinate, the gas phase temperature after the oxidation tail gas heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation tail gas heat exchanger, or the opening of the extraction tower top boundary position control regulating valve, etc. In the embodiment of the present invention, if the hydrogen peroxide concentration of the raffinate is to be detected, the current data of other sites other than the hydrogen peroxide concentration of the raffinate may be detected first.

[0035] Step S12, calculating the raffinate hydrogen peroxide concentration of the hydrogen peroxide device according to the current data of the other sites and based on a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by training a neural network using the fluctuation historical data of the raffinate hydrogen peroxide concentration and historical data of other sites related to the fluctuation historical data of the raffinate hydrogen peroxide concentration, and the fluctuation historical data is historical data in which the duration of abnormal operation within a preset time period is greater than a preset percentage of the preset time period.

[0036] Specifically, the embodiment of the present invention is based on the hydrogen peroxide concentration prediction model, and substitutes the current data of other detected sites to calculate the raffinate hydrogen peroxide concentration. The hydrogen peroxide concentration prediction model is obtained by training the neural network with the fluctuation historical data of the raffinate hydrogen peroxide concentration and the historical data of other sites related to the fluctuation historical data of the raffinate hydrogen peroxide concentration. The fluctuation historical data refers to the historical data in which the duration of abnormal operation data within the preset time is greater than the preset percentage of the preset time, and the preset percentage can be 30%, but is not limited to this. For example, assuming that the site has a total of 3 months of historical data, and the preset time is 1 month, then all historical data can be divided into three parts. In the historical data of the first month, 10 days of historical data are abnormal operation data (abnormal operation can be marked on the data), then 10 days account for about 33% of 1 month, which is greater than the preset percentage, then the historical data of the first month is the fluctuation historical data. In the historical data of the second and third months, 5 days and 4 days of historical data are abnormal operation data, accounting for less than 30% of 1 month, then the historical data of the second and third months are not fluctuation historical data. The historical data of other locations related to the fluctuation historical data may be directly acquired, or may be acquired according to the acquisition method provided in the following embodiments of the present invention.

[0037] Embodiment 2

[0038] In this embodiment, what is different from the first embodiment is that a method for establishing a hydrogen peroxide concentration prediction model is mainly provided, especially a more detailed method for obtaining historical data of other locations related to the fluctuation historical data. Other calculation methods are similar to those in the first embodiment and will not be repeated here.

[0039] Specifically, the hydrogen peroxide concentration prediction model is obtained by:

[0040] Step S21, obtaining the fluctuation history data of the raffinate hydrogen peroxide concentration;

[0041] Specifically, the data acquired by the embodiment of the present invention may come from the operation data stored in the distributed control system (DCS) and the laboratory information management system (LIMS) record data, but not all the historical data of the site are acquired, but only the fluctuation historical data is acquired. The meaning of the fluctuation historical data is as described above.

[0042] Step S22, calculating the grayscale correlation coefficient between the fluctuation historical data and the historical data of each of the other sites at the same operating time;

[0043] Specifically, since the historical data of other sites may also have a certain impact on the historical data of hydrogen peroxide concentration, the embodiment of the present invention also needs to determine the correlation between the above-mentioned fluctuation historical data and the historical data of other sites at the same operating time. The same operating time means that if the fluctuation historical data is data from January 2020 (even specific to a few days, hours, minutes and seconds), then the historical data of other sites for calculating the grayscale correlation coefficient should also use data from January 2020. Since there may be many other sites, the historical data of each site corresponding to the operating time can calculate the grayscale correlation coefficient with the fluctuation historical data.

[0044] Step S23, extracting the fluctuation historical data and the historical data of other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold value related to the site type;

[0045] Specifically, if the grayscale correlation coefficient calculated between the historical data of the H site and the fluctuation historical data is greater than the correlation coefficient threshold, it means that the historical data of the H site is highly correlated with the fluctuation historical data and needs to be extracted, that is, on the basis of extracting the fluctuation historical data, the historical data of the H site also needs to be extracted. The correlation coefficient threshold varies according to the site to be detected. For example, if the site to be detected in the present invention is the residual hydrogen peroxide concentration, the correlation coefficient threshold may be 0.55.

[0046] Step S24, training is performed based on the extracted data and the neural network to obtain the hydrogen peroxide concentration prediction model.

[0047] Specifically, for example, using the MATLAB neural network toolbox, the number of nodes in the input layer of the neural network is the number of other sites, the middle layer is 7 layers, the number of nodes in each layer is 50, and the output layer is 1 node, corresponding to the output variable. 70% of the data samples are used to train the model, 15% of the data samples are used to test the model, and 15% of the data samples are used to verify the model.

[0048] Embodiment 3

[0049] In this embodiment, what is different from Embodiments 1 and 2 is that a more detailed method for calculating the grayscale correlation coefficient between the fluctuation historical data and the historical data of one site among other sites at the same operating time is mainly provided. Other calculation methods are similar to Embodiments 1 and 2 and will not be repeated here.

[0050] Specifically, the grayscale correlation coefficient is calculated by the following formula:

[0051] Among them, x0 is the historical data of fluctuation, x i is the historical data of the ith site among other sites at the same operating time, ρ is the preset coefficient, and k is the serial number of the historical data. That is to say, assuming that the first site among other sites is taken as an example, the first site will have a lot of historical data, and these historical data correspond to the operating time of the fluctuation historical data. For example, the fluctuation historical data is a total of 60 data obtained every 12 hours within 30 days, then the first site also uses a total of 60 data obtained every 12 hours within these 30 days, and can be regarded as having serial numbers 1-60, and the sorting is based on time order.

[0052] Embodiment 4

[0053] In this embodiment, the difference from Embodiments 1 to 3 is that the preferred preset time length, data interval, other sites related to the fluctuation history data calculated by the above formula, and the preferred hydrogen peroxide concentration prediction model obtained by the above method are mainly provided. Other extraction methods are similar to Embodiments 1 to 3 and will not be repeated here.

[0054] Specifically, with respect to the preset time length and data interval, for the detection of hydrogen peroxide concentration, the preset time length may be 10 days, and the data interval may be 4 hours.

[0055] The preferred sites related to the fluctuation history data, that is, other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold related to the site type, include: the gas phase temperature after the oxidation tail gas heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation tail gas heat exchanger, and the opening of the extraction tower top boundary position control regulating valve. The grayscale correlation coefficients of these four calculated by the above formula are:

[0056]

[0057] Based on this, the obtained hydrogen peroxide concentration prediction model is:

[0058] X(t)=0.2994·X1(t)+0.1012·X2(t)+0.6557·X3(t)+0.0224·X4(t)

[0059] Among them, X1 is the current gas phase temperature A after the oxidation tail gas heat exchanger, X2 is the current working liquid receiving tank level, X3 is the current gas phase temperature B after the oxidation tail gas heat exchanger, X4 is the current top boundary position of the extraction tower, and X is the concentration of raffinate hydrogen peroxide. The gas phase temperature A after the oxidation tail gas heat exchanger and the gas phase temperature B after the oxidation tail gas heat exchanger are the gas phase temperatures measured at different locations after the gas passes through the tail gas heat exchanger.

[0060] Although specific numerical values ​​are provided in the above embodiment, such numerical values ​​are only preferred numerical values ​​in the embodiment of the present invention, and other numerical values ​​may also be used, and the embodiment of the present invention is not limited thereto.

[0061] Embodiment 5

[0062] Figure 3 1 is a structural block diagram of a device for calculating the concentration of residual hydrogen peroxide in a hydrogen peroxide device according to an embodiment of the present invention. Figure 3 As shown, the calculation device includes: a detection unit 1 and a processing unit 2, wherein the detection unit 1 is used to detect current data of other sites other than the residual hydrogen peroxide concentration of the hydrogen peroxide device; the processing unit 2 is used to calculate the residual hydrogen peroxide concentration of the hydrogen peroxide device based on the current data of the other sites and a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by performing neural network training on the fluctuation historical data of the residual hydrogen peroxide concentration and the historical data of other sites related to the fluctuation historical data of the residual hydrogen peroxide concentration.

[0063] Preferably, the fluctuation historical data is historical data in which the duration of abnormal operation data within a preset time period is greater than a preset percentage of the preset time period.

[0064] Preferably, the hydrogen peroxide concentration prediction model is obtained by: acquiring the fluctuation history data of the raffinate hydrogen peroxide concentration; calculating the grayscale correlation coefficient between the fluctuation history data and the history data of each site at other sites at the same operating time; extracting the fluctuation history data and the history data of other sites whose grayscale correlation coefficient is greater than a correlation coefficient threshold value related to the site type; and training based on the extracted data and a neural network to obtain the hydrogen peroxide concentration prediction model.

[0065] Preferably, the grayscale correlation coefficient is calculated by the following formula:

[0066] Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

[0067] Preferably, other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold related to the site type include: the gas phase temperature after the oxidation exhaust heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation exhaust heat exchanger, and the opening of the extraction tower top boundary position control regulating valve.

[0068] Preferably, the neural network training uses MATLAB neural network.

[0069] Preferably, 70% of the historical data of fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data of fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

[0070] This embodiment provides a device for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device, which is similar to the above-mentioned embodiments 1 to 4 and will not be described in detail here.

[0071] The raffinate hydrogen peroxide concentration calculation device comprises a processor and a memory. The detection unit and the processing unit are stored in the memory as program units. The processor executes the program units stored in the memory to realize corresponding functions.

[0072] The processor includes a kernel, and the kernel calls the corresponding program unit from the memory. One or more kernels can be set, and the concentration of residual hydrogen peroxide can be detected by adjusting kernel parameters.

[0073] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0074] An embodiment of the present invention provides a storage medium on which a program is stored. When the program is executed by a processor, the method for calculating the concentration of raffinate hydrogen peroxide is implemented.

[0075] An embodiment of the present invention provides a processor, which is used to run a program, wherein the method for calculating the concentration of raffinate hydrogen peroxide is executed when the program is run.

[0076] An embodiment of the present invention provides a device, the device including a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, the following steps are implemented:

[0077] Detecting current data of other sites other than the raffinate hydrogen peroxide concentration of the hydrogen peroxide device; calculating the raffinate hydrogen peroxide concentration of the hydrogen peroxide device based on the current data of the other sites and a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by training a neural network using the fluctuation history data of the raffinate hydrogen peroxide concentration and the historical data of other sites related to the fluctuation history data of the raffinate hydrogen peroxide concentration.

[0078] The fluctuation historical data is historical data in which the duration of abnormally running data within a preset time period is greater than a preset percentage of the preset time period.

[0079] The hydrogen peroxide concentration prediction model is obtained in the following manner: obtaining the fluctuation history data of the raffinate hydrogen peroxide concentration; calculating the grayscale correlation coefficient between the fluctuation history data and the history data of each site at other sites at the same operating time; extracting the fluctuation history data and the history data of other sites whose grayscale correlation coefficient is greater than a correlation coefficient threshold value related to the site type; and training based on the extracted data and a neural network to obtain the hydrogen peroxide concentration prediction model.

[0080] The grayscale correlation coefficient is calculated by the following formula:

[0081] Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

[0082] Other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold related to the site type include: the gas phase temperature after the oxidation exhaust heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation exhaust heat exchanger, and the opening of the extraction tower top boundary control regulating valve.

[0083] The neural network training uses MATLAB neural network.

[0084] 70% of the historical data on fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data on fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

[0085] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0086] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing the following method steps:

[0087] Detecting current data of other sites other than the raffinate hydrogen peroxide concentration of the hydrogen peroxide device; calculating the raffinate hydrogen peroxide concentration of the hydrogen peroxide device based on the current data of the other sites and a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by training a neural network using the fluctuation history data of the raffinate hydrogen peroxide concentration and the historical data of other sites related to the fluctuation history data of the raffinate hydrogen peroxide concentration.

[0088] The fluctuation historical data is historical data in which the duration of abnormally running data within a preset time period is greater than a preset percentage of the preset time period.

[0089] The hydrogen peroxide concentration prediction model is obtained in the following manner: obtaining the fluctuation history data of the raffinate hydrogen peroxide concentration; calculating the grayscale correlation coefficient between the fluctuation history data and the history data of each site at other sites at the same operating time; extracting the fluctuation history data and the history data of other sites whose grayscale correlation coefficient is greater than a correlation coefficient threshold value related to the site type; and training based on the extracted data and a neural network to obtain the hydrogen peroxide concentration prediction model.

[0090] The grayscale correlation coefficient is calculated by the following formula:

[0091] Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

[0092] Other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold related to the site type include: the gas phase temperature after the oxidation exhaust heat exchanger, the liquid level of the working liquid receiving tank, the opening of the gas phase temperature control regulating valve after the oxidation exhaust heat exchanger, and the opening of the extraction tower top boundary control regulating valve.

[0093] The neural network training uses MATLAB neural network.

[0094] 70% of the historical data on fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data on fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

[0095] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0096] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0099] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0100] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0101] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0102] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0103] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device, characterized in that: The calculation method includes: Detecting current data of other sites other than the raffinate hydrogen peroxide concentration of the hydrogen peroxide device, wherein the other sites are different measurement objects of different chemical devices or different measurement objects in the same chemical device; According to the current data of the other sites, based on the hydrogen peroxide concentration prediction model, the raffinate hydrogen peroxide concentration of the hydrogen peroxide device is calculated, wherein the hydrogen peroxide concentration prediction model is a model obtained by performing neural network training on the fluctuation historical data of the raffinate hydrogen peroxide concentration and the historical data of other sites at the same operating time related to the fluctuation historical data of the raffinate hydrogen peroxide concentration, wherein the fluctuation historical data is historical data in which the duration of abnormal operation data within a preset time period is greater than a preset percentage of the preset time period, and the fluctuation historical data comes from the operation data stored in the distributed control system (DSC) and the data recorded in the laboratory information management system (LIMS).

2. The method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to claim 1, characterized in that: The hydrogen peroxide concentration prediction model is obtained by the following method: Obtaining historical data of fluctuations in the concentration of the raffinate hydrogen peroxide; Calculate the grayscale correlation coefficient between the fluctuation historical data and the historical data of each site at other sites at the same operating time; Extract the fluctuation historical data and the historical data of other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold value related to the site type; The hydrogen peroxide concentration prediction model is obtained based on the extracted data and the neural network training.

3. The method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to claim 2, characterized in that: The grayscale correlation coefficient is calculated by the following formula: Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

4. The method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to claim 2, characterized in that: Other sites where the grayscale correlation coefficient is greater than the correlation coefficient threshold value related to the site type include: The gas phase temperature after the oxidation tail gas heat exchanger, the liquid level of the working liquid receiving tank, the gas phase temperature after the oxidation tail gas heat exchanger control the opening of the regulating valve and the top boundary position of the extraction tower control the opening of the regulating valve.

5. The method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to claim 1, characterized in that: The neural network training uses a MATLAB neural network, the number of nodes in the input layer of the neural network is the number of other sites, the middle layer is 7 layers, and the output layer corresponds to the number of output variables.

6. The method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to claim 1, characterized in that: 70% of the historical data on fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data on fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

7. A device for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device, characterized in that: The computing device comprises: A detection unit and a processing unit, wherein: The detection unit is used to detect current data of other sites other than the concentration of the residual hydrogen peroxide of the hydrogen peroxide device, wherein the other sites are different measurement objects of different chemical devices or different measurement objects in the same chemical device; The processing unit is used to calculate the raffinate hydrogen peroxide concentration of the hydrogen peroxide device according to the current data of the other sites and based on a hydrogen peroxide concentration prediction model, wherein the hydrogen peroxide concentration prediction model is a model obtained by performing neural network training on the fluctuation historical data of the raffinate hydrogen peroxide concentration and the historical data of other sites at the same operating time that are related to the fluctuation historical data of the raffinate hydrogen peroxide concentration, wherein the fluctuation historical data is historical data in which the duration of abnormal operation data within a preset time period is greater than a preset percentage of the preset time period, and the fluctuation historical data comes from the operation data stored in a distributed control system (DSC) and the data recorded in a laboratory information management system (LIMS).

8. The device for calculating the concentration of raffinate hydrogen peroxide of a hydrogen peroxide device according to claim 7, characterized in that: The hydrogen peroxide concentration prediction model is obtained by the following method: Obtaining historical data of fluctuations in the concentration of the raffinate hydrogen peroxide; Calculate the grayscale correlation coefficient between the fluctuation historical data and the historical data of each site at other sites at the same operating time; Extract the fluctuation historical data and the historical data of other sites whose grayscale correlation coefficient is greater than the correlation coefficient threshold value related to the site type; The hydrogen peroxide concentration prediction model is obtained based on the extracted data and the neural network training.

9. The device for calculating the concentration of raffinate hydrogen peroxide of a hydrogen peroxide device according to claim 8, characterized in that: The grayscale correlation coefficient is calculated by the following formula: Among them, x0 is the historical data of fluctuation, x i is the historical data of the i-th site among other sites at the same operating time, ρ is the preset coefficient, and k is the sequence number of the historical data.

10. The device for calculating the concentration of raffinate hydrogen peroxide of a hydrogen peroxide device according to claim 8, characterized in that: Other sites where the grayscale correlation coefficient is greater than the correlation coefficient threshold value related to the site type include: The gas phase temperature after the oxidation tail gas heat exchanger, the liquid level of the working liquid receiving tank, the gas phase temperature after the oxidation tail gas heat exchanger control the opening of the regulating valve and the top boundary position of the extraction tower control the opening of the regulating valve.

11. The device for calculating the concentration of raffinate hydrogen peroxide of a hydrogen peroxide device according to claim 7, characterized in that: The neural network training uses a MATLAB neural network, the number of nodes in the input layer of the neural network is the number of other sites, the middle layer is 7 layers, and the output layer corresponds to the number of output variables.

12. The device for calculating the concentration of raffinate hydrogen peroxide of a hydrogen peroxide device according to claim 7, characterized in that: 70% of the historical data on fluctuations in the raffinate hydrogen peroxide concentration and historical data of other locations related to the historical data on fluctuations in the raffinate hydrogen peroxide concentration are used to train the hydrogen peroxide concentration prediction model, 15% of the data are used to test the hydrogen peroxide concentration prediction model, and 15% of the data are used to verify the hydrogen peroxide concentration prediction model.

13. A machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device as claimed in any one of claims 1 to 6.

14. A processor, characterized in that: Used to run a program, wherein the program, when run, is used to execute the method for calculating the concentration of raffinate hydrogen peroxide in a hydrogen peroxide device according to any one of claims 1 to 6.

Images