Method and system for intelligent control of industrial processes

Abstract

This application discloses an industrial process intelligent control method and system. The industrial process intelligent control method includes the steps of acquiring raw measurement data of a controlled object when a stop condition has not been reached, analyzing the raw measurement data to obtain analyzed measurement data, learning the analyzed measurement data using a deep learning model to determine a control scheme, and controlling the controlled object based on the control action amount obtained from the control scheme. According to this embodiment of the industrial process intelligent control method, the new-generation process control platform analyzes and determines the error between the set value of the controlled variable and the measured value of the controlled variable and various evaluation indexes of the controlled industrial process to achieve a control effect that meets requirements and expectations.

Description

[Technical field]

[0001] The present application relates to the field of industrial process control, and more particularly to an intelligent industrial process control method and system. [Background technology]

[0002] Process control has a wide range of applications in industrial production, including power, petroleum, chemical industry, metallurgy, etc. By controlling process parameters such as temperature, pressure, flow rate, liquid level, composition, concentration, etc., the efficiency of industrial processes can be improved, energy consumption can be reduced, and production volume can be increased. With the continuous progress of technological innovation, industrial production processes are developing towards greater energy saving and intelligence.

[0003] Rapid and accurate measurement of process parameters is a fundamental prerequisite for realizing industrial process control. Non-contact laser measurement technology, as a potential and forward-looking online analysis technology, has great advantages compared with existing detection technologies. High-precision non-contact laser measurement can meet the requirements of online accurate and rapid measurement in high-temperature and harsh environments, and its measurement data is the basis for building an integrated intelligent industrial production system, laying the foundation for the efficient integration and intelligent control of information systems and physical systems.

[0004] Therefore, with the development of laser measurement technologies such as computed tomography-tunable diode laser absorption spectroscopy (CT-TDLAS) and laser-induced breakdown spectroscopy (LIBS), in-situ, non-contact and high-temporal and spatial resolution online measurement of temperature and material composition in industrial processes is realized. The industrial process intelligent monitoring and control platform combining high-precision laser measurement and numerical simulation has become an important development direction for process monitoring and control in the field of industrial measurement and control, and has high practicality.

[0005] However, there are various problems in today's industrial process control. For example, problems that tend to occur in measurement are that only point measurement, not surface measurement, is feasible, non-online measurement, and not high speed. Problems that tend to occur in control are delays in control and adjustment due to defects in the measurement method, measurement data is required for verification and optimization in CFD simulation, and the accuracy of the simulation results is insufficient, and the amount of data is insufficient.

[0006] For example, the response speed of CFD simulation analysis is slow, the simulation time is too long, and the simulation results are not accurate enough, which leads to problems such as delayed adjustment and inaccurate simulation analysis, and indirectly leads to the insufficient accuracy of the parameters output by the process control platform, making it impossible to achieve accurate control. Summary of the Invention [Problem to be solved by the invention]

[0007] In view of the above problems, one embodiment of the present invention provides an industrial process intelligent control method and system to solve the problems existing in the prior art. [Means for solving the problem]

[0008] In order to solve the above problems, an embodiment of the present application discloses an industrial process intelligent control method, the method comprising: if a stopping condition has not been reached, acquiring raw measurement data of the controlled object; analyzing the raw measurement data to obtain analyzed measurement data; learning the analyzed measurement data by a deep learning model to determine a control scheme; and controlling the controlled object based on a control action amount obtained from the control scheme.

[0009] In order to solve the above problem, an embodiment of the present application further discloses a terminal device, the terminal device comprising: one or more processors; and one or more machine-readable media having instructions stored thereon that, when executed by the one or more processors, cause the terminal device to perform the method.

[0010] In order to solve the above problems, an embodiment of the present application further discloses an industrial process intelligent control system, the industrial process intelligent control system including: a measuring device, an analysis module, a process control platform and an automatic control device; the measurement device is for acquiring raw measurement data of a controlled object if a stop condition has not been reached; the analysis module is for analyzing the raw measurement data to obtain analyzed measurement data; The process control platform is for learning the analyzed measurement data through a deep learning model and determining a control scheme; The automatic control device is configured to control the controlled object based on a control action amount obtained from the control scheme.

[0011] An embodiment of the present application further discloses a terminal device, the terminal device comprising: one or more processors; and one or more machine-readable media having stored thereon instructions that, when executed by the one or more processors, cause the terminal device to perform the method.

[0012] An embodiment of the present application discloses one or more machine-readable media having stored thereon instructions that, when executed by one or more processors, cause a terminal device to perform the above-described method. Effect of the Invention

[0013] From the above, it can be seen that the embodiments of the present application have the following advantages:

[0014] According to the industrial process intelligent control method of the embodiment of the present invention, the new generation process control platform can analyze and judge the error between the set value of the controlled variable and the measured value of the controlled variable, and various evaluation indexes of the industrial process to be controlled, so as to obtain a control effect that meets requirements and expectations. In order to more clearly describe the technical solutions in the embodiments of the present invention or the prior art, the drawings necessary for describing the embodiments or the prior art will be briefly described below. Of course, the drawings described below are only some embodiments of the present invention, and those skilled in the art can further obtain other drawings based on these embodiments without requiring creative labor. [Brief description of the drawings]

[0015] [Figure 1] FIG. 1 is a block diagram of an industrial process intelligent control system according to one embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic diagram of a deep learning process 1 according to an embodiment of the present invention and the corresponding deep learning model architecture. [Diagram 3] FIG. 3 is a schematic diagram of a deep learning process 2 according to an embodiment of the present invention and the corresponding deep learning model architecture. [Figure 4] FIG. 4 is a flow chart of an industrial process intelligent control method according to an embodiment of the present invention. [Diagram 5] FIG. 5 is a block diagram of an intelligent control system in a thermal power plant according to an embodiment of the present invention. [Figure 6] FIG. 6 is a block diagram of an intelligent control system for a semiconductor industrial process according to an embodiment of the present invention. [Figure 7] FIG. 7 is a schematic block diagram of a terminal device for performing the method according to the invention. [Figure 8] FIG. 8 shows a schematic representation of a storage unit for holding or carrying program code implementing the method of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. It is clear that the described embodiments are not all the embodiments, but only some of the embodiments of the present application. All other embodiments that a person skilled in the art can obtain based on the embodiments of the present application belong to the protection scope of the present application. First Example

[0017] The first embodiment of the present invention provides an industrial process intelligent control system. Figure 1 is a block diagram of an industrial process intelligent control system according to an embodiment of the present invention. The industrial process intelligent control system according to the embodiment of the present invention has a wide range of applications, for example, it can be applied to measuring temperature field and concentration field by laser, and the industrial process intelligent control system is for controlling a controlled object, and includes a measurement device 10, an analysis module 20, a process control platform 30 and an automatic control device 40.

[0018] said measuring device 10 is for acquiring raw measurement data of a controlled object; said analysis module 20 for analyzing said raw measurement data to obtain analyzed measurement data; the process control platform 30 is for determining a control scheme according to the analyzed measurement data; The automatic control device 40 is for controlling the controlled object based on a control action amount obtained from the control scheme, If a stopping condition has not been reached, the measurement device 10 repeatedly performs the operation of measuring the raw measurement data.

[0019] The measurement device 10 is a laser measurement device that implements various laser measurement techniques, such as computed tomography-tunable semiconductor laser absorption spectroscopy technique, laser-induced breakdown spectroscopy technique, etc., and may be a conventional measurement device. The raw measurement data is data obtained by measurement. However, it is not limited to this device, and any other measurement device that meets the measurement requirements can be applied to the industrial process intelligent control system of the present invention. In the embodiment of the present invention, the laser measurement device measures the acquired raw measurement data in real time, and selects the analysis module 20 of the corresponding measurement method based on the measurement device and the measurement parameter requirements. The analysis module 20 may be an analysis model, which analyzes the raw measurement data of the measurement device collected in real time, and obtains the measurement data after analysis, that is, the process parameters of the industrial process that need to be obtained.

[0020] In the embodiment of the present invention, a measuring device required for the control object is constructed according to the industrial process characteristics and the industrial process parameters that need to be measured quickly and accurately. The above-mentioned industrial processes are, for example, industrial processes in fields such as thermal power plants, semiconductor industry, engines, gas turbines, and metallurgical industries that need to obtain process parameters such as temperature, pressure, flow rate, liquid level, composition, concentration, etc. The measurement data after analysis by the module 20 in the embodiment of the present invention are process parameters such as temperature, pressure, flow rate, liquid level, composition, concentration, etc. This will be described as an example below.

[0021] As shown in Fig. 1, the process control platform 30 learns the analyzed measurement data through a deep learning model 50, where the source of the analyzed measurement data is not limited to the laser measurement method, but also includes other process parameters obtained by other measurement methods. Optimize the CFD simulation module 70 of the controlled object according to the learning result to obtain optimized CFD data, establish a CFD database 60, and develop a CFD big data analysis platform.

[0022] The analyzed measurement data is combined with the existing CFD data to obtain optimized CFD data through deep learning of the deep learning model 50, and a control scheme is provided by the process control platform 30 based on the further optimized CFD data.

[0023] The automatic controller 40 operates actuators according to a control scheme provided by the process control platform 30 to adjust the control actions and further modify the controlled variables of the controlled object 100 . If the stopping condition is not reached, the measurement device 10 repeatedly measures the controlled object 100, and the process control platform 30 analyzes and judges according to the error between the set value of the controlled variable and the measured value of the controlled variable and various evaluation indexes of the controlled industrial process, obtains the process parameters that meet the conditions, and achieves the expected control effect.

[0024] The measurement and control of process parameters in an industrial process are in a dynamic adjustment state, and the process parameters can be monitored in real time, and whether the controlled object 100 needs to be controlled by the automatic control device 40 is analyzed according to the error between the set value of the controlled variable and the measured value of the controlled variable, and various evaluation indexes of the controlled industrial process.

[0025] The evaluation of the control process of an industrial process intelligent control system based on laser measurement can be analyzed from five aspects: 1. evaluation index of the laser measurement method itself, 2. evaluation index of the CFD simulation model itself, 3. error between the measurement result and the simulation result, 4. error between the measured value of the controlled variable and the set value of the controlled variable, and 5. comprehensive evaluation index such as efficiency and energy consumption of the industrial process. When each evaluation index, error and comprehensive evaluation index are globally optimal, the expected control effect is achieved. In some embodiments, the stopping condition may be reaching the expected control condition, reaching the production time, etc. When it is determined that the stopping condition is reached, the cycle measurement is stopped.

[0026] FIG. 1 shows two deep learning processes for the process control platform 30, which are deep learning process 1 and deep learning process 2, respectively, shown by broken lines. Optimizing the measurement data and CFD data under different operating conditions of the controlled object 100 and forming big data for the CFD database is deep learning process 1, and this learning process is a long-term process, data accumulation takes months or years, and the amount of data is very important. Optimizing the CFD data with the real-time measurement data and existing CFD data in the CFD database, and further providing a control scheme by the process control platform is deep learning process 2, and this learning process is a transient real-time process, and timeliness is very important.

[0027] The above two deep learning processes may be executed in different deep learning models, or may be executed in the same deep learning model. In the present invention, only the deep learning model 50 represents an execution mechanism for executing the two different deep learning processes 1 and 2. In other embodiments, for example, the deep learning process 2 may be executed in one deep learning model, and the deep learning process 1 may be executed in another deep learning model.

[0028] For the deep learning process 1, Figure 2 is a schematic diagram of the model architecture for the deep learning process 1. As shown in Figure 2, in the deep learning model, the input layer 1 is the measurement data at different times and the result error of the deep learning process 2, the input layer 2 is various parameters of the CFD model, and the output layer is the optimized CFD data calculated according to the CFD model. Based on the output CFD data, a CFD database is established and CFD big data is formed.

[0029] For the deep learning process 2, Fig. 3 is a schematic diagram of the model architecture for the deep learning process 2. As shown in Fig. 3, in the deep learning model, the input layer 1 is real-time measurement data, the input layer 2 is CFD data in an existing CFD database, which includes the results of simple algebraic operations on the existing CFD data, and the output layer is optimized CFD data calculated based on the existing CFD data, and a control scheme is further provided by the process control platform based on the optimized CFD data.

[0030] In an optional embodiment, the process control platform may further include a first database unit and a first deep learning model unit for performing the following operations, respectively: The first database unit is for storing a plurality of analyzed measurement data; The first deep learning model unit is for learning the stored plurality of analyzed measurement data by a first deep learning model, and outputting CFD data processed by the first deep learning model.

[0031] In an optional embodiment, the process control platform may further include a second database unit and a second deep learning model unit for performing the following operations, respectively: The second database unit is for storing the plurality of analyzed measurement data and the historical CFD data; The first deep learning model unit is for learning the stored plurality of analyzed measurement data and historical CFD data by the second deep learning model, and optimizing the CFD simulation module.

[0032] In an optional embodiment, the analysis module may further include an analysis model determination unit and an analysis unit for performing the following operations, respectively: The analytical model determination unit is for determining an analytical model based on the raw measurement data of the measurement device collected in real time; The analysis unit is for analyzing according to the determined analysis model and obtaining post-analysis measurement data.

[0033] As can be seen from the above, the industrial process intelligent control system according to the first embodiment of the present invention has at least the following technical effects:

[0034] According to the industrial process intelligent control system of the embodiment of the present invention, the new generation process control platform can analyze and judge the error between the set value of the controlled variable and the measured value of the controlled variable, and various evaluation indexes of the controlled industrial process, so as to obtain a control effect that meets requirements and expectations.

[0035] In addition, the industrial process intelligent control system according to this embodiment further has at least the following advantages.

[0036] According to the embodiment of the industrial process intelligent control system of the present invention, inside the process control platform, the analyzed measurement data is learned by an artificial intelligence algorithm such as deep learning, where the source of the analyzed measurement data is not limited to the laser measurement method, but also includes other process parameters obtained by other measurement methods. The CFD simulation module of the controlled object is optimized according to the learning result to obtain optimized CFD data, a CFD database is established, and a CFD big data analysis platform is developed. The analyzed real-time measurement data and the existing CFD data are combined to obtain optimized CFD data by the deep learning of the process control platform, and a control scheme is further provided by the process control platform based on the optimized CFD data. The automatic control device operates the actuator according to the control scheme provided by the process control platform to adjust the control action amount and further change the controlled variable of the controlled object. The controlled object is repeatedly measured by the measuring device, and the process control platform analyzes and judges according to the error between the set value of the controlled variable and the measured value of the controlled variable and various evaluation indexes of the controlled industrial process to obtain the process parameters that meet the conditions and achieve the expected control effect.

[0037] The measurement and control of process parameters in an industrial process are in a dynamic adjustment state, and the process parameters can be monitored in real time, and whether the controlled object needs to be controlled by an automatic control device is analyzed according to the error between the set value of the controlled variable and the measured value of the controlled variable, and various evaluation indexes of the controlled industrial process.

[0038] The control system and method of the present invention are not only suitable for thermal power plants and semiconductor industrial processes, but can also be applied to fields such as engines, gas turbines and metallurgical industries to realize intelligent control of industrial processes, but are not limited to these fields.

[0039] The present invention establishes a CFD simulation model of a controlled object, calculates CFD data under different operating conditions of the controlled object, and provides the calculated CFD data in the form of a database to a process control platform. Inside the process control platform, the analyzed measurement data is accumulated and learned by an artificial intelligence algorithm such as deep learning, and the CFD simulation module of the controlled object is optimized according to the learning result to obtain optimized CFD data. Based on the CFD data under different operating conditions of the controlled object and the CFD data optimized by deep learning, a CFD database is established, and a CFD big data analysis platform is developed in the process control platform. The real-time process parameters of the controlled object are in a dynamic measurement state, and the CFD data optimized by deep learning is also in a dynamic adjustment state, and the established CFD database is constantly updated and improved. In the process control platform, data analysis is performed by an artificial intelligence algorithm such as deep learning between the analyzed measurement data, CFD data, and optimized CFD data. Second Example

[0040] A second embodiment of the present invention provides an intelligent control method for an industrial process. Figure 4 is a flow chart of steps of the intelligent control method for an industrial process according to the second embodiment of the present invention. As shown in Figure 4, the intelligent control method for an industrial process according to the embodiment of the present invention includes the following steps:

[0041] S101, if the stopping condition is not reached, raw measurement data of the control object is acquired.

[0042] In this step, the execution mechanism is, for example, a measurement device 10 of an industrial process intelligent control system, and the measurement device 10 is, for example, a laser measurement device implementing various laser measurement techniques such as computed tomography-tunable semiconductor laser absorption spectroscopy technology, laser induced breakdown spectroscopy technology, etc., and may be a conventional measurement device, but is not limited to this device, and any other measurement device that meets the measurement requirements can be applied to the industrial process intelligent control method of the present invention. The raw measurement data is, for example, data obtained by measurement.

[0043] The evaluation of the control process of an industrial process intelligent control system based on laser measurement can be analyzed from five aspects: 1. evaluation index of the laser measurement method itself, 2. evaluation index of the CFD simulation model itself, 3. error between the measurement result and the simulation result, 4. error between the measured value of the controlled variable and the set value of the controlled variable, and 5. comprehensive evaluation index such as efficiency and energy consumption of the industrial process. When each evaluation index, error and comprehensive evaluation index are globally optimal, the expected control effect is achieved. In some embodiments, the stopping condition may be reaching the expected control condition, reaching the production time, etc. When it is determined that the stopping condition is reached, the cycle measurement is stopped.

[0044] S102, the raw measurement data is analyzed to obtain analyzed measurement data.

[0045] In this step, the execution mechanism is, for example, the analysis module 20 of the industrial process intelligent control system. The analysis module 20 may be an analysis model, which analyzes the raw measurement data of the measurement device collected in real time, and obtains the analyzed measurement data, that is, the process parameters of the industrial process that need to be obtained.

[0046] S103, the deep learning model learns the analyzed measurement data and determines the control scheme.

[0047] The subject that performs this step is, for example, the process control platform 30 of an industrial process intelligent control system. The process control platform 30 learns the analyzed measurement data through the deep learning model 50, optimizes the analyzed measurement data through the deep learning model, and generates a control scheme according to the optimized CFD data.

[0048] S104, controlling the controlled object based on the control action amount obtained from the control scheme.

[0049] In this step, the execution mechanism is, for example, the automatic control device 40 of the industrial process intelligent control system. The automatic control device 40 operates the actuator according to the control scheme provided by the process control platform 30 to adjust the control action amount and further change the controlled variable of the controlled object 100.

[0050] As can be seen from the above, the industrial process intelligent control method according to the second embodiment of the present invention has at least the following technical effects:

[0051] According to the industrial process intelligent control method of the embodiment of the present invention, the new generation process control platform can analyze and judge the error between the set value of the controlled variable and the measured value of the controlled variable, and various evaluation indexes of the controlled industrial process, so as to obtain a control effect that meets requirements and expectations.

[0052] In addition, the industrial process intelligent control method according to this embodiment further has at least the following advantages.

[0053] According to the embodiment of the industrial process intelligent control method of the present invention, inside the process control platform, the analyzed measurement data is learned by an artificial intelligence algorithm such as deep learning, where the source of the analyzed measurement data is not limited to the laser measurement method, but also includes other process parameters obtained by other measurement methods. According to the learning result, the CFD simulation module of the controlled object is optimized to obtain optimized CFD data, a CFD database is established, and a CFD big data analysis platform is developed. The analyzed real-time measurement data is combined with the existing CFD data to obtain optimized CFD data by the deep learning of the process control platform, and a control scheme is further provided by the process control platform based on the optimized CFD data. The automatic control device operates the actuator according to the control scheme provided by the process control platform to adjust the control action amount and further change the controlled variable of the controlled object. The controlled object is repeatedly measured by the measuring device, and the process control platform analyzes and judges according to the error between the set value of the controlled variable and the measured value of the controlled variable and various evaluation indexes of the controlled industrial process to obtain the process parameters that meet the conditions and achieve the expected control effect.

[0054] The measurement and control of process parameters in an industrial process are in a dynamic adjustment state, and the process parameters can be monitored in real time, and whether the controlled object needs to be controlled by an automatic control device is analyzed according to the error between the set value of the controlled variable and the measured value of the controlled variable, and various evaluation indexes of the controlled industrial process.

[0055] The control system and method of the present invention are not only suitable for thermal power plants and semiconductor industrial processes, but can also be applied to fields such as engines, gas turbines and metallurgical industries to realize intelligent control of industrial processes, but are not limited to these fields. Third Example

[0056] FIG. 5 is a block diagram of an intelligent control system in a thermal power plant according to an embodiment of the present invention. As shown in FIG. 5, the boiler control of a thermal power plant is taken as an example to specifically describe an intelligent monitoring and control system and method combining laser measurement of temperature field and composition concentration field of an industrial process with numerical simulation. By using computer tomography-tunable semiconductor laser absorption spectroscopy technology to measure the temperature distribution and concentration distribution of other components in the combustion chamber and tail flue of the boiler, the application of high-end technologies such as big data, Internet of Things, cloud platform, etc. to the thermal power plant is realized, thereby improving the efficiency of the thermal power plant, realizing energy saving and emission reduction of the thermal power plant, and realizing the process intelligent control of the thermal power plant.

[0057] The system shown in Figure 5 mainly includes two parts: a measurement control system and an intelligent monitoring and control platform. In the measurement and control system part, a CT-TDLAS measurement device is built based on the combustion chamber structure and tail flue structure of the boiler to respectively measure the internal temperature field and gas composition concentration field of the combustion chamber, the temperature field of the tail flue, and the concentration fields of nitrogen oxides and ammonia before and after the denitrification device. The raw measurement data of each measurement device is collected in real time, and the raw measurement data is analyzed by the analysis module to obtain the temperature field and gas concentration field of the measurement cross section of the combustion chamber and tail flue, and the measurement data after analysis can be displayed on the display. The control scheme is provided through the analysis of the new generation process control platform. The powder supply amount and secondary air volume of the boiler are adjusted by the automatic control device so that the temperature and gas concentration of the combustion chamber and tail flue meet the set value and achieve the expected control effect.

[0058] In the intelligent monitoring and control platform part, the measurement data of CT-TDLAS is represented by ai, and a CT-TDLAS database Ai is established from the historical measurement data of CT-TDLAS. Based on the boiler structure, a CFD simulation model of the boiler combustion process under different operating conditions is established. By setting the CFD model parameters, the simulation results of CFD are obtained, such as the CFD data of the temperature field and concentration field, and a CFD database Di is established. Based on the laser optical path structure of the CT-TDLAS measurement device, the average value, fluctuation value and probability density function on each path of the temperature and concentration distribution of the CT-TDLAS data and the CFD data are obtained, respectively, and represented by Bi and B^i, and the CT-TDLAS measurement result path statistical value database Bi and the CFD simulation result path statistical value database B^i are compared. A correction function is constructed by the CFD model parameter database Ci and the CFD database Di, and a label is constructed by the difference value between the measurement result path statistical value database Bi and the simulation result path statistical value database B^i, and the processes of data preprocessing, feature selection, model construction, parameter optimization and evaluation are carried out by deep learning methods. If the model error after deep learning is equal to or greater than the set value error ε, continue optimizing the model parameters in the deep learning process.If the model error after deep learning is less than the set value error ε, output the CFD model parameters with the smallest error between the simulation result and the measurement result, update and improve the established model parameter database Ci to obtain the optimized CFD data, and further update and improve the established CFD database Di.

[0059] The intelligent monitoring and control platform part contains two kinds of deep learning processes: long-term accumulation process and transient real-time process. The long-term accumulation process is to obtain a CFD database through the accumulation of measurement data and CFD simulation data to form CFD big data. The transient real-time process is to obtain optimized CFD data based on real-time measurement data and existing CFD data to provide a control scheme bi.

[0060] In this example, the evaluation of the thermal power plant intelligent control system based on CT-TDLAS laser measurement in the control process is based on the model error analysis measurement results and simulation results in the deep learning process, in addition to the thermal power plant intelligent control system is based on the Evaluation of the CT-TDLAS measurement method, including the accuracy of TDLAS measured temperatures and concentrations, and the reconstruction accuracy of the CT algorithm, CFD simulation errors such as grid-scale errors, time-step errors, iteration errors, and input parameter errors; Errors between the measured values ​​and the set values, such as the internal combustion chamber temperature and gas composition concentration, the tail flue temperature, and the concentrations of nitrogen oxides and ammonia before and after the denitrification device; Power plant thermoeconomic indicators such as steam consumption and heat consumption of gas turbine units, heat consumption of power plants, coal consumption and power generation efficiency of power plants, etc. It can also be analyzed and evaluated from the aspects such as. When each evaluation index, error and comprehensive evaluation index are globally optimal, the expected control effect is achieved. Fourth embodiment

[0061] 6 is a block diagram of an intelligent control system for a semiconductor industrial process according to an embodiment of the present invention. As shown in FIG. 6, the control of a film deposition device in a semiconductor manufacturing process is taken as an example to specifically describe an intelligent monitoring and control system and method combining laser measurement and numerical simulation of the temperature field and composition concentration field of an industrial process. The purpose of controlling the performance of semiconductor materials is achieved by using computer tomography-tunable semiconductor laser absorption spectroscopy technology to measure the temperature field and gas concentration field distribution in the film deposition device.

[0062] However, the control system and method of the present invention are not only suitable for thermal power plants and semiconductor industrial processes, but can also be popularized in the fields of engines, gas turbines, metallurgical industries, etc., so as to realize intelligent control of industrial processes. However, it is not limited to these fields. It should be noted that the laser measurement technology adopted in the present invention not only includes computer tomography-tunable semiconductor laser absorption spectroscopy technology, but also includes laser induced breakdown spectroscopy technology, etc.

[0063] 7 is a schematic diagram of a hardware structure of a terminal device according to an embodiment of the present application. As shown in FIG. 7, the terminal device may include an input device 90, a processor 91, an output device 92, a memory 93, and at least one communication bus 94. The communication bus 94 is for realizing communication connections between elements. The memory 93 may include a high-speed RAM memory, and may further include a non-volatile memory NVM, such as at least one disk memory, and various programs may be stored in the memory 93 to complete various processing functions and realize steps of the method of the present embodiment.

[0064] Optionally, the processor 91 may be implemented by, for example, a central processing unit (abbreviated as CPU), an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic element, and the processor 91 is coupled to the input device 90 and the output device 92 by a wired or wireless connection.

[0065] Optionally, the input device 90 may include multiple input devices, such as a user interface for a user, a device interface for a device, a software programmable interface, a camera, and / or a sensor. Optionally, the device interface for a device may be a wired interface for data transmission between devices, or a hardware insertion interface for data transmission between devices (e.g., USB interface, serial port, etc.). Optionally, the user interface for a user may be, for example, a control button for a user, a voice input device for receiving voice input, and a touch-sensitive device for receiving user touch input (e.g., a touch screen with touch-sensitive function, touch pad, etc.). Optionally, the software programmable interface may be, for example, an entry for editing or modifying a user program, such as an input pin interface or an input interface of a chip. An audio input device, such as a microphone, may receive audio data. The output device 92 may include an output device, such as a display, an audio device, etc.

[0066] In this embodiment, the processor of the terminal device has the function of executing each module of the data processing apparatus in each device, and the specific functions and technical effects may be referred to the above embodiment and will not be repeated here.

[0067] Fig. 8 is a schematic diagram of a hardware structure of a terminal device according to another embodiment of the present application. Fig. 8 is a specific embodiment of the implementation process of Fig. 7. As shown in Fig. 8, the terminal device of this embodiment includes a processor 101 and a memory 102.

[0068] The processor 101 is for executing computer program codes stored in the memory 102 to realize the intelligent industrial process control method of FIG. 4 in the above embodiment.

[0069] The memory 102 is arranged to store various types of data to support the operation of the terminal device. Examples of such data include instructions for any application or method of operating on the terminal device, such as messages, photos, videos, etc. The memory 102 may include a random access memory (abbreviated as RAM) and may further include a non-volatile memory, such as at least one disk memory.

[0070] Optionally, the processor 101 is installed in the processing component 100. The terminal device may further include a communication component 103, a power component 104, a multimedia component 105, an audio component 106, an input / output interface 107, and / or a sensor component 108. The specific components included in the terminal device are set according to actual needs, and the embodiment is not limited thereto.

[0071] The processing component 100 typically controls the overall operation of the terminal device. The processing component 100 may include one or more processors 101 for executing instructions to complete all or some of the steps of FIG. 4 above. It should be noted that the processing component 100 may include one or more modules to facilitate interaction between the processing component 100 and other components. For example, the processing component 100 may include a multimedia module to facilitate interaction between the multimedia component 105 and the processing component 100.

[0072] The power component 104 provides power to the various components of the terminal device and may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the terminal device.

[0073] The multimedia component 105 includes a display screen that provides an output interface between the terminal device and the user. In some embodiments, the display screen may include a liquid crystal display (LCD) and a touch panel (TP). When the display screen includes a touch panel, the display screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to detect touches, slides, and gestures on the touch panel. The touch sensors can detect the boundaries of a touch or slide motion as well as the duration and pressure associated with the touch or slide operation.

[0074] The audio component 106 is arranged to output and / or input audio signals. For example, the audio component 106 includes one micro (MIC). When the terminal device is in an operation mode, such as a voice recognition mode, the micro is arranged to receive an external audio signal. The received audio signal can be further stored in the memory 102 or transmitted via the communication component 103. In some embodiments, the audio component 106 further includes a speaker for outputting the audio signal.

[0075] The input / output interface 107 provides an interface between the processing component 100 and a peripheral interface module, which may be a click wheel, buttons, etc. These buttons may include, but are not limited to, volume buttons, a start button, and a lock button.

[0076] The sensor component 108 includes one or more sensors for providing various state assessments to the terminal device. For example, the sensor component 108 can detect the start / stop state of the terminal device, the relative position of the components, and the presence or absence of contact between the user and the terminal device. The sensor component 108 may include a proximity sensor arranged to detect the presence of a nearby object in the absence of physical contact, including detecting the distance between the user and the terminal device. In some embodiments, the sensor component 108 may further include a camera, etc.

[0077] The communication component 103 is arranged to facilitate wired or wireless communication between the terminal device and other devices. The terminal device may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In one embodiment, the terminal device may include a SIM card slot for inserting a SIM card, so that the terminal device can log into a GPRS network and establish communication with a server over the Internet.

[0078] From the above, it can be seen that the communication component 103, the audio component 106, the input / output interface 107, and the sensor component 108 in the embodiment of FIG. 8 can all be implementations of the input devices in the embodiment of FIG.

[0079] An embodiment of the present application provides a terminal device, the terminal device including one or more processors and one or more machine-readable media having stored thereon instructions that, when executed by the one or more processors, cause the terminal device to perform a method according to one or more of the embodiments of the present application.

[0080] Each embodiment in this specification is described in a step-by-step manner, with each embodiment focusing on the differences from other embodiments, and the same or similar parts between each embodiment can be referenced to each other.

[0081] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have acquired the basic, inventive concept. Therefore, it is intended that the appended claims be interpreted as including all changes and modifications that are within the scope of the preferred embodiments and the present application.

[0082] Finally, in this specification, relational terms such as first and second are merely intended to distinguish one entity or operation from another and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Terms such as "comprises," "comprises," or other variations thereof are intended to cover non-exclusive inclusions, such that a process, method, article, or terminal device that includes a set of elements not only includes those elements, but also includes other elements not expressly listed or that are inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined in the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0083] The above is a detailed description of the method and system for intelligent control of industrial processes according to the present application. Although the present application uses specific examples to explain the principles and embodiments of the present application, the description of the above examples is only intended to help understand the method and main idea of ​​the present application. Those skilled in the art can make changes to the specific embodiments and application scope based on the idea of ​​the present application. As mentioned above, the contents of the present application should not be construed as limitations of the present application.

Claims

1. if the stopping condition has not been reached, acquiring raw measurement data of the controlled object by a measurement device; analyzing the raw measurement data to obtain analyzed first measurement data; a step of inputting the first measurement data into a deep learning model trained using, as teacher data, a plurality of second measurement data measured and analyzed at a time different from that of the first measurement data and a plurality of CFD data of temperature fields and concentration fields corresponding to the plurality of second measurement data, thereby acquiring optimal CFD data of the temperature field and the concentration field output from the deep learning model, and determining a control scheme based on the acquired optimal CFD data; and controlling the controlled object based on a control action amount obtained from the control scheme.

2. A step of storing the plurality of second measurement data and the plurality of CFD data corresponding to the plurality of second measurement data in a database; and using the plurality of second measurement data and the plurality of CFD data stored in the database as teacher data to train the deep learning model.

2. The method for intelligent control of an industrial process according to claim 1, further comprising:

3. The step of analyzing the raw measurement data to obtain analyzed first measurement data includes: determining an analytical model based on raw measurement data of the measurement device collected in real time; 2. The method for intelligent control of an industrial process according to claim 1, further comprising: obtaining measurement data after analysis according to the determined analysis model.

4. 2. The method of claim 1, wherein the raw measurement data includes at least one of temperature, pressure, flow rate, liquid level, composition, and concentration.

5. 2. The method of claim 1, wherein the measurement device comprises a laser measurement device.

6. 2. The method for intelligently controlling an industrial process according to claim 1, wherein the measuring device is a laser measuring device of computed tomography-tunable semiconductor laser absorption spectroscopy.

7. 2. The method of claim 1, wherein the stop condition includes either the end of a production time or the achievement of a control goal.

8. 1. An industrial process intelligent control system for controlling a control target, comprising: The industrial process intelligent control system includes a measuring device, an analysis module, a process control platform and an automatic control device; the measurement device is for acquiring raw measurement data of a controlled object if a stop condition has not been reached; the analysis module is for analyzing the raw measurement data to obtain analyzed first measurement data; the process control platform is for inputting the first measurement data into a deep learning model trained using, as teacher data, a plurality of second measurement data measured and analyzed at a time different from that of the first measurement data and a plurality of CFD data of temperature fields and concentration fields corresponding to the plurality of second measurement data, thereby acquiring optimal CFD data of temperature fields and concentration fields output from the deep learning model, and determining a control scheme based on the acquired optimal CFD data; 11. An industrial process intelligent control system, comprising: an automatic control device for controlling the controlled object based on a control action amount obtained from the control scheme.

9. The process control platform includes: a database unit for storing the plurality of second measurement data and the plurality of CFD data corresponding to the plurality of second measurement data; The industrial process intelligent control system according to claim 8, further comprising: a deep learning model unit configured to train the deep learning model using the plurality of second measurement data and the plurality of CFD data stored in the database unit as teacher data.

10. The analysis module includes: an analytical model determination unit for determining an analytical model based on raw measurement data of the measurement device collected in real time; 9. The industrial process intelligent control system according to claim 8, further comprising: an analysis unit for analyzing according to the determined analysis model and obtaining measurement data after analysis.

11. 9. The industrial process intelligent control system of claim 8, wherein the raw measurement data includes at least one of temperature, pressure, flow rate, liquid level, composition, and concentration.

12. 10. The industrial process intelligent control system of claim 9, wherein the measurement device comprises a laser measurement device.

13. 9. The industrial process intelligent control system according to claim 8, wherein the measuring device is a laser measuring device based on computed tomography-tunable semiconductor laser absorption spectroscopy technology or a laser measuring device based on laser induced breakdown spectroscopy technology.

14. 9. The industrial process intelligent control system of claim 8, wherein the stop condition includes either an end of production time or an achievement of a control objective.

15. A terminal device, comprising: one or more processors; and one or more machine-readable media on which are stored instructions which, when executed by the one or more processors, cause the terminal device to perform the methods according to one or more of claims 1-8.

16. One or more machine-readable media having stored thereon instructions which, when executed by one or more processors, cause a terminal device to perform the method according to one or more of claims 1-8.

17. A program which, when executed by one or more processors, causes a terminal device to perform the methods described in one or more of claims 1-8.

Images