Fault sample generation method and device of Hypol process, equipment and medium

Abstract

The invention relates to a fault sample generation method and device of a Hypol process, equipment and a medium, in particular to the field of polypropylene production.The method comprises the steps that a dynamic mathematical model based on the polypropylene process of a full-mixed-flow reactor is established; unknown parameters of the dynamic mathematical model are identified through a chaos particle swarm optimization algorithm, a dynamic mechanism model is obtained, and then typical fault sample data of the polypropylene process are obtained according to the dynamic mechanism model; pre-training the typical fault sample data of the polypropylene process by using a generative adversarial neural network to obtain a pre-training model; and inputting fault sample data of actual production into the pre-training model to obtain fault sample data of the corresponding production line. According to the method provided by the invention, the mechanism model is utilized to generate enough fault data of a specified type, pre-training is carried out, finally, the model is finely adjusted by using actual fault data, a fault sample generator for generating the specified fault type is obtained, and specified fault data of different types can be generated.

Description

Technical Field

[0001] The present invention relates to the field of polypropylene production, and in particular to a method and device, equipment and medium for generating a fault sample of a Hypol process, and in particular to a method and device, equipment and medium for generating a fault sample of polypropylene produced by a Hypol process. Background Art

[0002] Polypropylene is an important chemical product with a wide range of applications and an important position in various fields. However, its process mechanism is complex, the working conditions are changeable, and the production process requires high-precision parameters as a basis. Once a failure occurs, some control variables will exceed the control range, which will easily lead to production safety accidents such as explosions. Therefore, real-time monitoring and diagnosis of polypropylene process failures is particularly important.

[0003] At present, the method of diagnosing polypropylene process faults often starts from the device, and the type of equipment fault is judged by analyzing the equipment vibration spectrum. However, for the process industry, not only the mechanical equipment itself may fail, but the process itself may also fail due to certain factors during material transportation, heat transfer, mass transfer and reaction, which is difficult to detect through the equipment vibration. Starting from the perspective of process production data, this problem can be effectively solved through machine learning and other means. However, due to the particularity of the process industry, it is generally difficult to collect enough fault data to train the machine learning model, and due to the complexity of the process, it is difficult to establish a strict simulation model to study the fault data.

[0004] For example, CN101046687A discloses a melt index detection fault diagnosis system for industrial polypropylene production, which includes a field intelligent instrument connected to an industrial process object, a DCS system and a host computer, wherein the DCS system is composed of a data interface, a control station and a database; the intelligent instrument, the DCS system and the host computer are connected in sequence, and the host computer includes a standardization processing module, a wavelet decomposition module, a principal component analysis function module, a wavelet reconstruction function module, a support vector machine classifier function module and a fault judgment module.

[0005] In summary, the prior art only analyzes the melt index detection failure, which still has certain limitations and cannot meet the needs of diagnosing other failures, and is not conducive to monitoring and diagnosing various failures in polypropylene production. Summary of the invention

[0006] In view of the problems existing in the prior art, the purpose of the present invention is to provide a method and device, equipment and medium for generating fault samples of a Hypol process, so as to solve the defect that the prior art cannot effectively monitor and diagnose various faults in polypropylene production.

[0007] To achieve this object, the present invention adopts the following technical solutions:

[0008] In a first aspect, the present invention provides a method for generating a fault sample of a Hypol process, the method comprising:

[0009] S1. Establish a dynamic mathematical model of polypropylene process based on full mixed flow reactor;

[0010] S2. Identify the unknown parameters of the dynamic mathematical model through the chaotic particle swarm optimization algorithm to obtain the dynamic mechanism model, and then obtain the typical fault sample data of the polypropylene process according to the dynamic mechanism model;

[0011] S3, using a generative adversarial neural network to pre-train typical failure sample data of the polypropylene process to obtain a pre-training model;

[0012] S4. Input the fault sample data of actual production into the pre-training model to obtain the fault sample data of the corresponding production line.

[0013] The fault sample generation method provided by the present invention generates sufficient fault data of specified types using a mechanism model according to the mechanism of polypropylene process pipeline faults, heat transfer faults, mass transfer faults and reaction faults, performs pre-training through a generative adversarial network, and finally fine-tunes the model using a small amount of fault data in actual production, and finally obtains a fault sample generator that can generate specified fault types. This method can not only generate different types of specified fault data without restriction, but also has higher accuracy, a wider range of generated data, and is closer to actual production data than general mechanism models.

[0014] As a preferred technical solution of the present invention, the dynamic mathematical model includes: material balance relationship, energy balance relationship and melt index.

[0015] As a preferred technical solution of the present invention, the material balance relationship includes a relationship formula of slurry solid content F, a relationship formula of hydrogen concentration and a relationship formula of propylene monomer concentration Cp.

[0016] Preferably, the slurry solid content F is expressed as follows:

[0017]

[0018] R p =α p F c C p V l

[0019] Where, F is the volume fraction of the solid part in the slurry, %; t is the time, min; I c is the catalyst feeding rate, kg / min; R pis the propylene polymerization rate, kg / min; W is the discharge rate, m 3 / min; V l is the slurry volume, m 3 ρ s is the polymer particle density, kg / m 3 ;f c is the mass fraction of catalyst in the polymer, %; I p is the propylene monomer feed rate, kg / min; ρ p is the density of propylene, kg / m 3 ρ c is the catalyst density; ρ pp is the density of polypropylene, kg / m 3 ; α p is the reaction constant related to the rate of propylene consumption, kg / (mol·min).

[0020] Preferably, the hydrogen concentration relationship is as follows:

[0021]

[0022] In the formula, is the hydrogen concentration in the gas phase, %; is the hydrogen feed rate, kg / min; is a constant related to the hydrogen consumption rate, kg / min; M H is the relative molecular mass of hydrogen, kg / mol; is a constant related to hydrogen dissolution; P is the pressure in the reactor, Pa; V g is the volume of the gas phase in the reactor, m 3 ; C p is the propylene monomer concentration, mol / m 3 ; V l is the volume of the liquid phase in the reactor, m 3 .

[0023] Preferably, the propylene monomer concentration Cp is expressed as follows:

[0024]

[0025] C p＝ 1 / V s

[0026] Where V s is the molecular volume of saturated propylene liquid at temperature T, m 3 ; R is the gas constant, J / (K·mol); T c is the critical temperature of propylene, K; P c is the critical pressure of propylene, Pa; ZRA is an empirical parameter, T r is the comparative temperature of propylene.

[0027] Preferably, the energy balance relationship includes a reaction temperature relationship, a circulating water outlet temperature relationship and a cooling water outlet temperature relationship.

[0028] Preferably, the reaction temperature relationship is as follows (8):

[0029]

[0030] In the formula, ρ p,g is the density of gas phase propylene, kg / m 3 ; C m is the average thermal melting point of the gas phase; ρ p,l is the density of liquid propylene, kg / m 3 ; C p,l Liquid propylene hot melt, J / (kg·K); C p,s is the mixed solid hot melt, J / (kg·K); T is the reaction temperature, K; -ΔH r is the heat of polymerization of propylene, J / mol; M p is the molecular weight of propylene, kg / mol; T p,i is the feed temperature, K; T r is the reference temperature, K; K j is the jacket heat transfer coefficient, J / (min·m 2 ·K); A is the heat transfer area of ​​the jacket, m 2 ; T j,o is the outlet temperature of jacket circulating water, K; G v is the steam flow rate, mol / min; k is the mass fraction of steam to the condenser; ΔH vm is the vaporization enthalpy of propylene, J / mol; T re is the condenser heat flow outlet temperature, K.

[0031] Preferably, the circulating water outlet temperature relationship is as follows:

[0032]

[0033] Where V j is the jacket volume, m 3 ρ j is the jacket cooling water density, kg / m 3 ; C j For jacket hot melt, J / (kg·K); I j is the jacket water flow rate, kg / m 3 ; T j,in is the jacket cold flow inlet temperature, K.

[0034] Preferably, the cooling water outlet temperature relationship is as follows:

[0035]

[0036] In the formula, ρ l is the slurry density, kg / m 3 ; C l For slurry hot melt, J / (kg·K); I l is the slurry flow rate, kg / min; T l,o is the condenser cooling water outlet temperature, K; T l,i is the condenser cooling water inlet temperature, K.

[0037] Preferably, the melt index M fr As follows:

[0038] M fr =M wc -3.36 ×10 18.56

[0039]

[0040] In the formula, K p is the propylene polymerization constant; γ S is the adsorption factor of propylene monomer on the active center of the catalyst; is the chain transfer rate to hydrogen 1 / min; M WC is the cumulative molecule; Q is the polydispersity index of polypropylene.

[0041] As a preferred technical solution of the present invention, the chaotic particle swarm optimization algorithm contains Tent chaotic mapping.

[0042] Preferably, the model parameters to be identified in the identification include: a reaction constant α related to the propylene rate consumption p , a constant related to the hydrogen consumption rate Constants related to hydrogen dissolution Empirical parameter Z RA , Propylene polymerization constant K p , chain transfer rate to hydrogen

[0043] Preferably, the input parameters in the identification include the catalyst feed rate I c , propylene monomer feed rate I p , hydrogen feed rate Jacket water flow I j 、Slurry flow I l , jacket cold flow inlet temperature T j,in and the condenser cooling water inlet temperature T l,i, output parameters include melt index M fr .

[0044] As a preferred technical solution of the present invention, the method of obtaining typical fault sample data of the polypropylene process based on the dynamic mechanism model includes: simulating typical fault data in the polypropylene process using the dynamic mechanism model to obtain typical fault sample data of the polypropylene process.

[0045] As a preferred technical solution of the present invention, the typical fault sample data of the polypropylene process is normalized before the pre-training.

[0046] As a preferred technical solution of the present invention, the pre-training includes inputting typical polypropylene process fault sample data and fault type labels as information into a generator and a discriminator to obtain a pre-training model.

[0047] Preferably, the objective function used in the pre-training is as follows:

[0048]

[0049] In the formula, G is the generator, D is the discriminator, and P data Represents the initial given data distribution, P z represents the randomly initialized noise distribution, G(z) represents the input is the random seed z and the output is the virtual fault data, D(f(x)) represents the input is the mixed operation data generated by the mechanism model and the parameter identification module, and the output is the probability.

[0050] In a second aspect, the present invention provides a fault sample generating device for a Hypol process, the fault sample generating device comprising:

[0051] Dynamic mathematical model acquisition module, used to establish a dynamic mathematical model of polypropylene process based on full mixed flow reactor;

[0052] The dynamic mechanism model acquisition model is used to identify the unknown parameters of the dynamic mathematical model through the chaotic particle swarm optimization algorithm to obtain the dynamic mechanism model;

[0053] Typical fault sample data acquisition model, used to obtain typical fault sample data of polypropylene process based on dynamic mechanism model;

[0054] A pre-training model acquisition module is used to pre-train typical fault sample data of polypropylene process using a generative adversarial neural network to obtain a pre-training model;

[0055] The corresponding production line fault sample data acquisition module is used to input the actual production fault sample data into the pre-training model to obtain the fault sample data of the corresponding production line.

[0056] In a third aspect, the present invention provides an electronic device, the electronic device comprising:

[0057] at least one processor; and a memory communicatively coupled to the at least one processor;

[0058] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the fault sample generation method of the Hypol process described in the first aspect.

[0059] In a fourth aspect, the present invention provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and when the computer executable instructions are executed by a processor, the method for generating fault samples of the Hypol process described in the first aspect is implemented.

[0060] Compared with the prior art solutions, the present invention has the following beneficial effects:

[0061] The present invention identifies the empirical parameters in the simulation model through an optimization algorithm, so that the data generated by the simulation model is closer to the real production data, and then pre-trains the conditional generative adversarial network with the fault data in the simulation model, and then fine-tunes the conditional generative adversarial network with a small amount of real production data, and finally obtains virtual production data that can generate specified fault types, which greatly facilitates subsequent fault warning work. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flow chart of a method for generating a fault sample of a Hypol process provided by the present invention;

[0063] Figure 2 It is a schematic diagram of a fault sample generating device of a Hypol process provided by the present invention;

[0064] Figure 3 is a schematic diagram of an electronic device provided by the present invention;

[0065] Figure 4 1 is a comparison chart of the data results obtained in Example 1 of the present invention.

[0066] In the figure: 10 - electronic device, 11 - processor, 12 - ROM, 13 - RAM, 14 - bus, 15 - I / O interface, 16 - input unit, 17 - output unit, 18 - storage unit, 19 - communication unit.

[0067] The present invention is further described in detail below. However, the following examples are only simplified examples of the present invention and do not represent or limit the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the claims. DETAILED DESCRIPTION

[0068] To better illustrate the present invention and facilitate understanding of the technical solution of the present invention, typical but non-limiting embodiments of the present invention are as follows:

[0069] This embodiment provides a method for generating a fault sample of a Hypol process, such as Figure 1 As shown, the method includes:

[0070] S1. Establish a dynamic mathematical model of polypropylene process based on full mixed flow reactor;

[0071] S2. Identify the unknown parameters of the dynamic mathematical model through the chaotic particle swarm optimization algorithm to obtain the dynamic mechanism model, and then obtain the typical fault sample data of the polypropylene process according to the dynamic mechanism model;

[0072] S3, using a generative adversarial neural network to pre-train typical failure sample data of the polypropylene process to obtain a pre-training model;

[0073] S4. Input the fault sample data of actual production into the pre-training model to obtain the fault sample data of the corresponding production line.

[0074] Wherein, the dynamic mathematical model includes: material balance relationship, energy balance relationship and melt index.

[0075] The material balance relationship includes a relationship between slurry solid content F, a relationship between hydrogen concentration and a relationship between propylene monomer concentration Cp.

[0076] Wherein, the relationship between the slurry solid content F is as follows:

[0077]

[0078]

[0079] Where, F is the volume fraction of the solid part in the slurry, %; t is the time, min; I c is the catalyst feeding rate, kg / min; R p is the propylene polymerization rate, kg / min; W is the discharge rate, m 3 / min; V l is the slurry volume, m 3 ρ s is the polymer particle density, kg / m 3 ;f c is the mass fraction of catalyst in the polymer, %; I p is the propylene monomer feed rate, kg / min; ρ p is the density of propylene, kg / m 3 ρc is the catalyst density; ρ pp is the density of polypropylene, kg / m 3 ; α p is the reaction constant related to the rate of propylene consumption, kg / (mol·min).

[0080] Wherein, the hydrogen concentration relationship is as follows (5):

[0081]

[0082] In the formula, is the hydrogen concentration in the gas phase, %; is the hydrogen feed rate, kg / min; is a constant related to the hydrogen consumption rate, kg / min; M H is the relative molecular mass of hydrogen, kg / mol; is a constant related to hydrogen dissolution; P is the pressure in the reactor, Pa; V g is the volume of the gas phase in the reactor, m 3 ; C p is the propylene monomer concentration, mol / m 3 ; V l is the volume of the liquid phase in the reactor, m 3 .

[0083] Wherein, the propylene monomer concentration Cp is expressed by the following equations (6) and (7):

[0084]

[0085] Where V s is the molecular volume of saturated propylene liquid at temperature T, m 3 , R is the gas constant, J / (K·mol); T c is the critical temperature of propylene, K; P c is the critical pressure of propylene, Pa; Z RA is an empirical parameter, T r is the comparative temperature of propylene.

[0086] Wherein, the energy balance relationship includes a reaction temperature relationship, a circulating water outlet temperature relationship and a cooling water outlet temperature relationship.

[0087] Wherein, the reaction temperature relationship is as follows (8):

[0088]

[0089] In the formula, ρ p,g is the density of gas phase propylene, kg / m 3 , C mis the average thermal melting point of the gas phase, ρ p,l is the density of liquid propylene, kg / m 3 , C p,l Liquid propylene hot melt, J / (kg·K); C p,s is the mixed solid hot melt, J / (kg·K); T is the reaction temperature, K; -ΔH r is the heat of polymerization of propylene, J / mol; M p is the molecular weight of propylene, kg / mol; T p,i is the feed temperature, K; T r is the reference temperature, K; K j is the jacket heat transfer coefficient, J / (min·m 2 ·K); A is the heat transfer area of ​​the jacket, m 2 ; T j,o is the outlet temperature of jacket circulating water, K; G v is the steam flow rate, mol / min; k is the mass fraction of steam to the condenser, ΔH vm is the vaporization enthalpy of propylene, J / mol; T re is the condenser heat flow outlet temperature, K.

[0090] Wherein, the circulating water outlet temperature relationship is as follows (9):

[0091]

[0092] Where V j is the jacket volume, m 3 ρ j Jacket cooling water density, kg / m 3 ; C j For jacket hot melt, J / (kg·K); I j is the jacket water flow rate, kg / m 3 ; T j,in is the jacket cold flow inlet temperature, K.

[0093] The cooling water outlet temperature relationship is as follows (10):

[0094]

[0095] In the formula, ρ l is the slurry density, kg / m 3 ; C l For slurry hot melt, J / (kg·K); I l is the slurry flow rate, kg / min; T l,o is the condenser cooling water outlet temperature, K; T l,i is the condenser cooling water inlet temperature, K.

[0096] Wherein, the melt index M fr As shown in formula (11) to formula (13):

[0097] M fr =M wc -3.36 ×10 18.56 (11)

[0098]

[0099]

[0100] In the formula, K p is the propylene polymerization constant; γ S is the adsorption factor of propylene monomer on the active center of the catalyst; is the chain transfer rate to hydrogen, 1 / min; M WC is the cumulative molecule; Q is the polydispersity index of polypropylene.

[0101] Wherein, the chaotic particle swarm optimization algorithm contains Tent chaotic mapping.

[0102] The formula of the Tent chaotic mapping is as follows (14):

[0103]

[0104] In the formula, α is taken as 0.499.

[0105] In the present invention, the purpose of identification is to use a chaotic particle swarm optimization algorithm to identify and determine the unknown parameters in the dynamic mathematical model, obtain the unknown parameters, and then obtain a dynamic mechanism model.

[0106] Exemplarily, the process of the chaotic particle swarm optimization algorithm is as follows:

[0107] Step 1: Initialize particle swarm parameters.

[0108] Among them, the particle swarm parameters include particle swarm size, particle dimension, number of iterations, inertia weight, learning factor and iteration step range.

[0109] Step 2: Initialize the speed and position of each particle through Tent chaos mapping;

[0110] Among them, the position of the particle represents the variable to be identified.

[0111] Step 3: Calculate the fitness of each particle using the mean square error between the output value of the dynamic mathematical model and the real historical value;

[0112] Step 4: Update the individual historical optimal fitness value and position of each particle, as well as the group historical optimal value and position;

[0113] Step 5: Determine whether the end condition is met (the error between the model output value and the true value under the same conditions is ≤5%, or the specified number of iterations is reached, ≥200 times). If it is met, the task is terminated and the value of the identification parameter is output, otherwise it continues;

[0114] Step 6: Update the speed and position of each particle, and determine whether the speed and position of the particle are out of range. If so, correct it through Tent chaotic mapping, otherwise return to Step 3.

[0115] The model parameters to be identified in the identification include: the reaction constant α related to the propylene rate consumption p , a constant related to the hydrogen consumption rate Constants related to hydrogen dissolution Empirical parameter Z RA , Propylene polymerization constant K p , chain transfer rate to hydrogen

[0116] The input parameters in the identification include the catalyst feed rate I c , propylene monomer feed rate I p , hydrogen feed rate Jacket water flow I j 、Slurry flow I l , jacket cold flow inlet temperature T j,in and the condenser cooling water inlet temperature T l,i , output parameters include melt index M fr .

[0117] Wherein, obtaining typical fault sample data of the polypropylene process according to the dynamic mechanism model includes: simulating typical fault data in the polypropylene process using the dynamic mechanism model to obtain typical fault sample data of the polypropylene process.

[0118] Wherein, before the pre-training, the typical fault sample data of the polypropylene process is normalized.

[0119] In the present invention, the normalization process may be performed by using a normalization method commonly used in the art, such as a minimum-maximum normalization method, to normalize the data.

[0120] The pre-training includes: inputting typical polypropylene process fault sample data and fault type labels as information into a generator and a discriminator to obtain a pre-training model;

[0121] The objective function used in the pre-training is as follows:

[0122]

[0123] In the formula, G is the generator, D is the discriminator, and P data Represents the initial given data distribution, P z represents the randomly initialized noise distribution, G(z) represents the input is the random seed z and the output is the virtual fault data, D(f(x)) represents the input is the mixed operation data generated by the mechanism model and the parameter identification module, and the output is the probability.

[0124] Among them, the generator used in the pre-training uses a randomly generated seed as input to generate data of the same dimension as the process variable data of the polypropylene simulation model, and uses the call discriminator and the comparison loss of the discriminator to update the generator. At this time, the discriminator is not updated.

[0125] Among them, the discriminator in the pre-training is used to accept a pair of data from the generator and process variable data generated by the polypropylene simulation model, and mark the data from the generator as 0 and the data from the mixed operation as 1, use the discriminator to classify them and compare them with the labels, and update the discriminator through the loss.

[0126] Among them, the function valuations of the generator and the discriminator in the pre-training are opposite.

[0127] In the present invention, the process of acquiring the fault sample data of the corresponding production line includes: inputting the fault data and fault sample labels generated in the actual production process as input information into the pre-training model to obtain the fault data of the corresponding production line.

[0128] Furthermore, this embodiment provides a fault sample generating device for a Hypol process, such as Figure 2 As shown, the device comprises:

[0129] Dynamic mathematical model acquisition module, used to establish a dynamic mathematical model of polypropylene process based on full mixed flow reactor;

[0130] The dynamic mechanism model acquisition model is used to identify the unknown parameters of the dynamic mathematical model through the chaotic particle swarm optimization algorithm to obtain the dynamic mechanism model;

[0131] Typical fault sample data acquisition model, used to obtain typical fault sample data of polypropylene process based on dynamic mechanism model;

[0132] A pre-training model acquisition module is used to pre-train typical fault sample data of polypropylene process using a generative adversarial neural network to obtain a pre-training model;

[0133] The corresponding production line fault sample data acquisition module is used to input the actual production fault sample data into the pre-training model to obtain the fault sample data of the corresponding production line.

[0134] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0135] Further, the present embodiment provides an electronic device, which 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 processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0136] like Figure 3 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The I / O interface 15 is also connected to the bus 14.

[0137] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0138] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the fault sample generation method of the Hypol process.

[0139] In some embodiments, the fault sample generation method of the Hypol process may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the fault sample generation method of the Hypol process described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the fault sample generation method of the Hypol process in any other appropriate manner (e.g., by means of firmware).

[0140] Various implementations of the systems and techniques described above 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), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including 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.

[0141] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0142] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0144] The systems and techniques described herein may be implemented in a computing system that includes backend 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 frontend 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 backend components, middleware components, or frontend components. The components of the system may 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), a blockchain network, and the Internet.

[0145] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0146] The server provided in this embodiment includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for generating fault samples of the Hypol process described above is implemented.

[0147] Unless otherwise specifically stated, terms such as processing, computing, calculating, determining, displaying, etc. may refer to the actions and / or processes of one or more processing or computing systems, or similar devices, which operate and convert data represented as physical (e.g., electronic) quantities within registers or memories of a processing system into other data similarly represented as physical quantities within memories, registers, or other such information storage, transmission, or display devices of the processing system. Information and signals may be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.

[0148] Those skilled in the art will also appreciate that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments of the present invention may all be implemented as electronic hardware, computer software, or a combination thereof. In order to clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described around their functions. Whether such functions are implemented as hardware or software depends on the specific application and the design constraints imposed on the entire system. A skilled person may implement the described functions in an alternative manner for each specific application, but such implementation decisions should not be interpreted as departing from the scope of protection of the present disclosure.

[0149] The steps of the method or algorithm described in conjunction with the embodiments herein may be directly embodied as hardware, a software module executed by a processor, or a combination thereof. The software module may be located in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a mobile disk, a CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The ASIC may be located in a user terminal. Of course, the processor and the storage medium may also be present in a user terminal as discrete components.

[0150] For software implementation, the techniques described in the present invention can be implemented with modules (e.g., procedures, functions, etc.) that perform the functions described in the present application. These software codes can be stored in a memory unit and executed by a processor. The memory unit can be implemented within the processor or outside the processor. In the latter case, it is coupled to the processor in a communication manner via various means, which are well known in the art.

[0151] Further, in order to illustrate the fault sample generation effect that can be achieved by the fault sample generation method of the Hypol process provided by the present invention, the following actual example is used for exemplary description, as follows:

[0152] Example 1

[0153] This example uses the monitoring variable 1 in the normal production data, fault 1 data, and fault 2 data in the Hypol process of a polypropylene plant to conduct experiments, as follows:

[0154] 1. First, the chaotic particle swarm algorithm is used to randomly generate the data to be identified. After the operating condition data (model input data) of the normal historical data is input, the difference between the output data of the model and the process data and the real data is found as the condition of fitness evaluation and returned to the chaotic particle swarm algorithm. The chaotic particle swarm algorithm is used to optimize the data to be identified.

[0155] 2. Put the identified data into the dynamic mathematical model as known parameters, and finally form a dynamic mechanism model. By analyzing the mechanism of fault 1 and fault 2, set relevant input data, input it into the dynamic model, and generate simulated fault data;

[0156] 3. After obtaining the simulated fault data, the simulated fault data is pre-trained by using a generative adversarial neural network to obtain a pre-trained model;

[0157] 4. Finally, the fault data and fault sample labels generated in the actual production process are input into the pre-training model as input information to obtain the fault sample data of the corresponding production line and realize the generation of fault sample data.

[0158] The final result is Figure 4 As shown in the figure, it can be seen that the virtual data and actual data obtained by the fault sample generation method provided by the present invention have a good match, and various fault samples can be efficiently generated, which is beneficial to the detection of faults in polypropylene production.

[0159] The preferred embodiments of the present invention are described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0160] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0161] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A method for generating a fault sample of a Hypol process, characterized in that: The fault sample generation method comprises: S1. Establish a dynamic mathematical model of polypropylene process based on full mixed flow reactor; S2. Identify the unknown parameters of the dynamic mathematical model through the chaotic particle swarm optimization algorithm to obtain the dynamic mechanism model, and then obtain the typical fault sample data of the polypropylene process according to the dynamic mechanism model; S3, using a generative adversarial neural network to pre-train typical failure sample data of the polypropylene process to obtain a pre-training model; S4. Input the fault sample data of actual production into the pre-training model to obtain the fault sample data of the corresponding production line.

2. The method for generating a fault sample according to claim 1, characterized in that: The dynamic mathematical model includes: material balance relationship, energy balance relationship and melt index.

3. The fault sample generation method according to claim 2, characterized in that: The material balance relationship includes a relationship between the slurry solid content F, a relationship between the hydrogen concentration, and a relationship between the propylene monomer concentration Cp; Preferably, the slurry solid content F is expressed as follows: R p =α p F c C p V l Where, F is the volume fraction of the solid part in the slurry, %; t is time, min; I c is the catalyst feeding rate, kg / min; R p is the propylene polymerization rate, kg / min; W is the discharge rate, m 3 / min; V l is the slurry volume, m 3 ; ρ s is the polymer particle density, kg / m 3 ;f c is the mass fraction of catalyst in the polymer, %; I p is the propylene monomer feed rate, kg / min; ρ p is the density of propylene, kg / m 3 ; ρ c is the catalyst density; ρ pp is the density of polypropylene, kg / m 3 ; α p is the reaction constant related to the propylene rate consumption, kg / (mol·min); Preferably, the hydrogen concentration relationship is as follows: In the formula, is the hydrogen concentration in the gas phase, %; is the hydrogen feed rate, kg / min; is a constant related to the hydrogen consumption rate, kg / min; M H is the relative molecular mass of hydrogen, kg / mol; is a constant related to hydrogen dissolution; P is the pressure in the reactor, Pa; V g is the volume of the gas phase in the reactor, m 3 ; C p is the propylene monomer concentration, mol / m 3 ; V l is the volume of the liquid phase in the reactor, m 3 ; Preferably, the propylene monomer concentration Cp is expressed as follows: C p =1 / V s Where V s is the molecular volume of saturated propylene liquid at temperature T, m 3 , R is the gas constant, J / (K·mol); T c is the critical temperature of propylene, K; P c is the critical pressure of propylene, Pa; Z RA is an empirical parameter, T r is the contrast temperature of propylene; Preferably, the energy balance relationship includes a reaction temperature relationship, a circulating water outlet temperature relationship, and a cooling water outlet temperature relationship; Preferably, the reaction temperature relationship is as follows: In the formula, ρ p,g is the density of gas phase propylene, kg / m 3 ; C m is the average thermal melting point of the gas phase; ρ p,l is the density of liquid propylene, kg / m 3 ; C p,l Liquid propylene hot melt, J / (kg·K); C p,s is the mixed solid hot melt, J / (kg·K); T is the reaction temperature, K; -ΔH r is the heat of polymerization of propylene, J / mol; M p is the molecular weight of propylene, kg / mol; T p,i is the feed temperature, K; T r is the reference temperature, K; K j is the jacket heat transfer coefficient, J / (min·m 2 ·K); A is the heat transfer area of ​​the jacket, m 2 ; T j,o is the outlet temperature of jacket circulating water, K; G v is the steam flow rate, mol / min; k is the mass fraction of steam to the condenser, ΔH vm is the vaporization enthalpy of propylene, J / mol; T re is the condenser heat flow outlet temperature, K; Preferably, the circulating water outlet temperature relationship is as follows: Where V j is the jacket volume, m 3 ; ρ j Jacket cooling water density, kg / m 3 ; C j For jacket hot melt, J / (kg·K); I j is the jacket water flow rate, kg / min; T j,in is the jacket cold flow inlet temperature, K; Preferably, the cooling water outlet temperature relationship is as follows: In the formula, ρ l is the slurry density, kg / m 3 ; C l For slurry hot melt, J / (kg·K); I l is the slurry flow rate, kg / min; T l,o is the cooling water outlet temperature of the condenser, K; T l,i is the condenser cooling water inlet temperature, K; Preferably, the melt index M fr As follows: M fr =M wc -3.36 ×10 18.56 In the formula, K p is the propylene polymerization constant; γ S is the adsorption factor of propylene monomer on the active center of the catalyst; is the chain transfer rate to hydrogen, 1 / min; M WC is the cumulative molecule; Q is the polydispersity index of polypropylene.

4. The method for generating a fault sample according to claim 1, wherein: The chaotic particle swarm optimization algorithm contains Tent chaotic mapping; Preferably, the model parameters to be identified in the identification include: a reaction constant α related to the propylene rate consumption p , a constant related to the hydrogen consumption rate Constants related to hydrogen dissolution Empirical parameter Z RA , Propylene polymerization constant K p , chain transfer rate to hydrogen Preferably, the input parameters in the identification include the catalyst feed rate I c , propylene monomer feed rate I p , hydrogen feed rate Jacket water flow I j 、Slurry flow I l , jacket cold flow inlet temperature T j,in and the condenser cooling water inlet temperature T l,i , output parameters include melt index M fr .

5. The fault sample generation method according to claim 1, characterized in that: The obtaining of typical fault sample data of the polypropylene process according to the dynamic mechanism model comprises: simulating typical fault data in the polypropylene process using the dynamic mechanism model to obtain typical fault sample data of the polypropylene process.

6. The method for generating a fault sample according to claim 1, characterized in that: Before the pre-training, the typical fault sample data of the polypropylene process is normalized.

7. The method for generating a fault sample according to claim 1, characterized in that: The pre-training includes inputting typical polypropylene process fault sample data and fault type labels as information into a generator and a discriminator to obtain a pre-training model; Preferably, the objective function used in the pre-training is as follows: In the formula, G is the generator, D is the discriminator, and P data Represents the initial given data distribution, P z represents the randomly initialized noise distribution, G(z) represents the input is the random seed z and the output is the virtual fault data, D(f(x)) represents the input is the mixed operation data generated by the mechanism model and the parameter identification module, and the output is the probability.

8. A fault sample generating device for a Hypol process, characterized in that: The fault sample generating device comprises: Dynamic mathematical model acquisition module, used to establish a dynamic mathematical model of polypropylene process based on full mixed flow reactor; The dynamic mechanism model acquisition model is used to identify the unknown parameters of the dynamic mathematical model through the chaotic particle swarm optimization algorithm to obtain the dynamic mechanism model; Typical fault sample data acquisition model, used to obtain typical fault sample data of polypropylene process based on dynamic mechanism model; A pre-training model acquisition module is used to pre-train typical fault sample data of polypropylene process using a generative adversarial neural network to obtain a pre-training model; The corresponding production line fault sample data acquisition module is used to input the actual production fault sample data into the pre-training model to obtain the fault sample data of the corresponding production line.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for generating a fault sample of the Hypol process according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer executable instructions, and when the computer executable instructions are executed by a processor, the method for generating a fault sample of a Hypol process according to any one of claims 1 to 7 is implemented.

Images