Process air conditioner control method and device based on prediction model

Abstract

The invention discloses a process air conditioner control method and device based on a prediction model. Real-time data of a working site where the process air conditioner is located are collected in real time, and the real-time data are preprocessed to generate standardized time sequence data; inputting the standardized time sequence data into a multilayer perceptron (MLP) model to generate process air conditioner control parameters under the current working condition; and when a preset evaluation triggering condition is met, performance evaluation is conducted on the control parameters through a GRU model, and when an evaluation result meets an expected control effect, the control parameters are transmitted to a PLC of the process air conditioner so that the control parameters of the process air conditioner can be adjusted. The control parameters of the process air conditioner are optimized through intelligent data processing and dynamic prediction, the problems that in the traditional industrial control technology, intelligence is insufficient, the data processing capacity is limited, and energy efficiency management is insufficient are solved, the efficiency, precision and adaptability of the process air conditioner are improved, energy consumption is reduced, and the self-adaptability and reliability of the process air conditioner are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment control, and in particular to a process air conditioning control method and device based on a prediction model. Background Art

[0002] In the blending and tobacco storage workshops during tobacco production, the temperature and humidity of the blending and tobacco storage rooms are kept stable by optimizing the process air conditioning control to ensure the quality of the tobacco produced. Conventional process air conditioning control technology consists of sensors, controllers, actuators, and communication networks. It collects various data in the production process and uses preset logic programs to adjust the operating status of the equipment. The transmission of data acquisition and control instructions relies on industrial communication protocols. At the same time, industrial control systems are usually combined with SCADA to provide equipment monitoring, alarm, data storage, and visualization functions to facilitate remote monitoring and management by operators.

[0003] Traditional air-conditioning control mainly relies on preset logic programs, which cannot be adaptively adjusted according to the complex and changeable production environment, cannot dynamically analyze and predict the operating status of equipment, and it is difficult to achieve intelligent optimization control; traditional industrial control cannot efficiently process and utilize large-scale real-time data, and can only control based on historical experience, and cannot perform complex data analysis, such as machine learning or deep learning technology.

[0004] Based on this, a method is needed to improve the intelligent control level of process air conditioning, so as to improve the efficiency, accuracy and adaptability of process air conditioning. Summary of the invention

[0005] The present invention provides a process air conditioning control method and device based on a prediction model to improve the intelligent control level of the process air conditioning, thereby improving the efficiency, accuracy and adaptability of the process air conditioning.

[0006] According to one aspect of the present invention, there is provided a process air conditioning control method based on a prediction model, comprising:

[0008] Real-time data of the operation site where the process air conditioner is located is collected in real time, and the real-time data is preprocessed to generate standardized time series data;

[0009] Inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions;

[0010] When the preset evaluation trigger conditions are met, the control parameters are performance evaluated through the gated recurrent unit GRU model. When the evaluation results meet the expected control effects, the control parameters are transmitted to the PLC of the process air conditioner to adjust the control parameters of the process air conditioner.

[0011] Optionally, the real-time collection of multi-source heterogeneous data at the work site where the process air conditioner is located, and preprocessing the multi-source heterogeneous data to generate standardized time series data, include:

[0012] Collect the real-time data from the air conditioning system, boiler system, refrigeration system and MES system of the work site, the real-time data including fresh air valve opening, return air valve opening, chilled water valve opening, heating valve opening, humidification valve opening, supply air valve opening, supply fan frequency, chilled water temperature and pressure, heating steam temperature and pressure, humidification steam temperature and pressure, fresh air temperature and humidity, return air temperature and humidity, supply air temperature and humidity, blending workshop sensor temperature and humidity and blending workshop air duct opening;

[0013] Processing missing values ​​and outliers in the real-time data and performing data standardization or normalization;

[0014] Data features are extracted based on correlation coefficient, moving average and time trend to generate the standardized time series data.

[0015] Optionally, the step of inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions includes:

[0016] The historical data of process air conditioning is used for model training to predict the output of process air conditioning, support the solution of optimization objectives, and generate control strategies.

[0017] Optionally, the performing performance evaluation on the control parameter by using a gated recurrent unit (GRU) model includes:

[0018] Receiving control parameters outputted from the MLP model and historical performance data of the process air conditioner, and conducting performance evaluation in terms of response speed, energy-saving effect and environmental adaptability, wherein the evaluation indicators are the real-time response time of the air conditioning system, the temperature and humidity control accuracy and the energy consumption optimization ratio;

[0019] Capture the dynamic changes and dependencies of input data based on recurrent neural networks;

[0020] As for the dynamic changes and dependencies, the real-time response time, temperature and humidity control accuracy, and energy consumption optimization ratio are quantitatively evaluated.

[0021] Optionally, the performing performance evaluation on the control parameter by using a gated recurrent unit (GRU) model includes:

[0022] Analyze the real-time operation data and historical feedback data of the process air conditioner to calculate the deviation between the actual performance and the target effect;

[0023] When the deviation is less than a preset range, it is determined that the evaluation result meets the expected control effect; if the deviation is not less than the preset range, the MLP model is self-optimized, and the self-optimization includes adjusting the parameters and training strategy of the MLP model according to the deviation and updating the model structure of the MLP model.

[0024] Optionally, the method further includes:

[0025] responding to an external intervention instruction, wherein the intervention instruction includes at least one control parameter of the process air conditioner;

[0026] The process air conditioning control parameters under the current working conditions are adjusted according to the control parameters.

[0027] Optionally, the trigger condition includes: reaching a preset evaluation time and / or the number of responses to the intervention instruction within a preset time period exceeds a preset value.

[0028] According to another aspect of the present invention, there is provided a process air conditioning control device based on a prediction model, comprising:

[0029] A data acquisition unit, used for real-time acquisition of real-time data of the operation site where the process air conditioner is located, and preprocessing the real-time data to generate standardized time series data;

[0030] A parameter generation unit, used for inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions;

[0031] The performance evaluation unit is used to perform performance evaluation on the control parameters through a gated recurrent unit (GRU) model when a preset evaluation trigger condition is met, and when the evaluation result meets the expected control effect, the control parameters are transmitted to the PLC of the process air conditioner to adjust the control parameters of the process air conditioner.

[0032] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0033] At least one processor; and a memory communicatively connected to the at least one processor; wherein 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 process air conditioning control method based on the prediction model described in any embodiment of the present invention.

[0034] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the process air conditioning control method based on the prediction model described in any embodiment of the present invention when executed.

[0035] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the process air conditioning control method based on the prediction model described in any embodiment of the present invention is implemented.

[0036] The technical solution of the embodiment of the present invention optimizes the control parameters of the process air conditioner through intelligent data processing and dynamic prediction, thereby solving the problems of insufficient intelligence, limited data processing capability and insufficient energy efficiency management in traditional industrial control technology, improving the efficiency, accuracy and adaptability of the process air conditioner, reducing energy consumption, and enhancing the adaptability and reliability of the process air conditioner.

[0037] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 is a flow chart of a process air conditioning control method based on a prediction model provided in Example 1 of the present invention;

[0040] Figure 2 is a structural schematic diagram of a process air conditioning control device based on a prediction model provided in Embodiment 2 of the present invention;

[0041] Figure 3 It is a structural schematic diagram of an electronic device for implementing a process air conditioning control method based on a prediction model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0043] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] Embodiment 1

[0045] Figure 1 This is a flowchart of a process air conditioning control method based on a prediction model provided in the first embodiment of the present invention. This embodiment is applicable to the case where the control parameters of the process air conditioning are adjusted. The method can be executed by a process air conditioning control device based on a prediction model. The process air conditioning control device based on a prediction model can be implemented in the form of hardware and / or software. The process air conditioning control device based on a prediction model can be configured in an electronic device. Figure 1 As shown, the method includes:

[0046] S110, collecting real-time data of the operation site where the process air conditioner is located in real time, and preprocessing the real-time data to generate standardized time series data.

[0047] In the production workshops of specific products, such as the blending and storage workshops in tobacco production, in order to ensure the quality of tobacco output, process air conditioning is required to keep the temperature and humidity of the blending room and the storage room in a stable state in real time. On this basis, other systems in the workshop also play a role in production, including: the refrigeration system that provides chilled water for the process air conditioning, the boiler system that provides heat energy for the air conditioning, and the MES system that confirms the production schedule. Temperature and humidity control is required during production, and temperature and humidity control is suspended during shutdown periods. Therefore, when collecting real-time data, it is necessary to collect data in real time from the air conditioning system, boiler system, refrigeration system, and MES system, and transmit data through industrial protocols.

[0048] In the embodiment of the present invention, S110 specifically includes the following steps:

[0049] Collect real-time data from the air conditioning system, boiler system, refrigeration system and MES system at the work site. The real-time data includes the opening of the fresh air valve, the opening of the return air valve, the opening of the chilled water valve, the opening of the heating valve, the opening of the humidification valve, the opening of the supply air valve, the frequency of the supply fan, the temperature and pressure of the chilled water, the temperature and pressure of the heating steam, the temperature and pressure of the humidification steam, the temperature and humidity of the fresh air, the temperature and humidity of the return air, the temperature and humidity of the supply air, the temperature and humidity of the sensor in the mixing workshop, and the opening of the air duct in the mixing workshop;

[0050] Process missing values ​​and outliers in real-time data and perform data standardization or normalization;

[0051] Data features are extracted based on correlation coefficients, moving averages, and time trends to generate standardized time series data.

[0052] Through ETL technology, data is collected from the above systems, including fresh air, return air, chilled water, heating, humidification, air supply valve opening, air supply fan frequency, and various temperature, humidity, and pressure data, as well as sensor temperature, humidity, and air duct opening in the blending workshop. These data are the original materials for subsequent analysis and decision-making of the system. In addition, data transmission is based on industrial protocols to ensure efficient data transmission.

[0053] The timestamp synchronization mechanism is used to integrate asynchronous data from multiple devices to avoid confusion and errors caused by data asynchrony, maintain data consistency, and enable the data to accurately reflect the actual operating status of the system.

[0054] Process missing values ​​and outliers in the data to ensure data integrity and accuracy; standardize or normalize the data to eliminate the impact of different dimensions and make the data comparable; extract data features based on correlation coefficients, moving averages, and time trends, explore the potential value of the data, and provide high-quality data input for subsequent modules.

[0055] S120, inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions.

[0056] The multi-layer perceptron (MLP) model is a data-driven nonlinear modeling method used to generate effective control strategies. MLP is a feedforward neural network consisting of an input layer, multiple hidden layers, and an output layer, with weights connecting the layers. In the process of controlling process air conditioning, the MLP model receives standardized time series data of processed data, including a variety of temperature and humidity data, valve opening data, etc. as input. When constructing the MLP model, the control environment data, including fresh air temperature and humidity, return air temperature and humidity, supply air temperature and humidity, sensor temperature and humidity in the mixing workshop; duct valve opening, historical data of duct opening in the mixing workshop are selected as input variables, and fresh air valve opening, return air valve opening, chilled water valve opening, heating valve opening, humidification valve opening, supply air valve opening, and supply fan frequency are selected as output variables.

[0057] In the embodiment of the present invention, S120 specifically includes the following steps: using historical data of the process air conditioner to perform model training, predict the output of the process air conditioner, and optimize the solution of the target to generate a control strategy.

[0058] By learning the input-output relationship of the system and using historical data to train the model, the model can automatically adjust the weights. For example, it can learn from historical data how to adjust the opening of each valve and the frequency of the blower to achieve the best control effect under different temperature and humidity environments. The trained model can predict the output of the process air conditioner in the future based on the current real-time data input, and then generate a control strategy.

[0059] The MLP model can handle multivariable coupling and complex constraints, such as temperature range limits and equipment power limits in process air conditioning operation. In the control task, it only outputs the optimal input value at the current moment, that is, it predicts the optimal control parameters of the current air conditioning equipment, adapts to dynamic environmental changes, and is widely used in complex industrial scenarios, effectively improving the intelligent control level of process air conditioning.

[0060] S130. When the preset evaluation trigger conditions are met, the performance of the control parameters is evaluated through the gated recurrent unit GRU model. When the evaluation results meet the expected control effects, the control parameters are transmitted to the PLC of the process air conditioner to adjust the control parameters of the process air conditioner.

[0061] The GRU model is a lightweight recurrent neural network that can effectively capture long-term dependencies in time series data. Compared with traditional RNNs, it solves the problem of gradient vanishing or gradient exploding by introducing a gating mechanism, making the model perform better when processing long sequence data.

[0062] The GRU model is used to evaluate the control scheme output by the MLP model. It will be considered from multiple key dimensions, including response speed, energy-saving effect, and environmental adaptability. For example, the control scheme is evaluated to see whether it can enable the air-conditioning system to quickly reach and stabilize within the set temperature and humidity range (response speed); whether it can minimize energy consumption while meeting environmental requirements (energy-saving effect); and whether it can adapt to different external environmental changes and fluctuations in production conditions within the workshop (environmental adaptability).

[0063] In an embodiment of the present invention, the performance evaluation of the control parameters is performed through a gated recurrent unit GRU model, including:

[0064] Analyze the real-time operation data and historical feedback data of process air conditioning and calculate the deviation between actual performance and target effect;

[0065] When the deviation is less than the preset range, it is determined that the evaluation result meets the expected control effect; if the deviation is not less than the preset range, the MLP model is self-optimized, and the self-optimization includes adjusting the parameters and training strategy of the MLP model according to the deviation and updating the model structure of the MLP model.

[0066] The evaluation results of the GRU model provide an important basis for deciding whether to use control parameters. If the evaluation results show that the current control scheme performs poorly in some aspects, the MLP model can be adjusted and optimized, or the user can intervene to make manual adjustments or trigger model retraining to ensure that the air-conditioning system is always in an efficient, stable and energy-saving operating state.

[0067] By automatically analyzing and adjusting the parameters of the MKP model, the control accuracy and responsiveness are continuously improved based on real-time operation data and historical feedback. Combined with machine learning technology, the model structure or training strategy is dynamically adjusted according to the gap between the actual performance of the system and the expected performance, thereby optimizing the control solution and improving the system efficiency and stability.

[0068] Specifically, the internal parameters of the MLP model are adjusted dynamically. For example, based on the feedback of real-time and historical data, the weights and biases of the model are optimized so that the model can more accurately reflect the relationship between input variables (such as ambient temperature and humidity, valve opening, etc.) and output variables (such as the optimal opening of each valve, the optimal frequency of the blower, etc.), thereby improving the accuracy of the prediction.

[0069] Adjust the training strategy of the MLP model, including learning rate, batch size, number of training rounds, etc. A suitable training strategy can make the model converge faster, improve training efficiency, avoid overfitting or underfitting problems, and enable the model to maintain good performance under different working conditions.

[0070] According to the operation status and performance requirements of the process air conditioner, the structure of the MLP model is optimized. For example, the number of hidden layers is increased or decreased, the number of neurons in the hidden layer is adjusted, etc., to adapt to control tasks of different complexity and improve the expression and generalization capabilities of the model.

[0071] When it is detected that there is a large deviation between the control effect and the expected target, such as a decrease in the temperature and humidity control accuracy of the process air conditioning, an abnormal increase in energy consumption, etc., the optimization program will be automatically started to adjust and optimize the MLP model.

[0072] In addition, the basis for data processing in self-optimization can be achieved through data analysis, which includes two parts: processing analysis and data analysis methods. In the mechanism analysis part, the basic data flow is constructed through data storage and reading operations, and the data trend is analyzed by model construction and simulation detection. The data analysis methods cover supervised learning (such as linear regression, decision trees, deep neural networks, etc.) and unsupervised learning (such as principal component analysis, cluster analysis, etc.), and can apply adaptive algorithms for different tasks to mine data features and rules. This module provides key technical support for subsequent model optimization and performance improvement.

[0073] In an embodiment of the present invention, the method may include the following steps:

[0074] responding to an external intervention instruction, wherein the intervention instruction includes at least one control parameter of the process air conditioner;

[0075] Adjust the process air conditioning control parameters under the current working conditions according to the control parameters.

[0076] When the effect of the control parameters does not meet expectations, the user is allowed to intervene in the system decision-making process. The user can evaluate and judge the operating status of the system and take corresponding measures to ensure the stable operation and control effect of the air-conditioning system. The user can decide to retrain the MLP model based on the actual situation. For example, when the layout of the production workshop changes or the equipment is upgraded, the original model may no longer be applicable. At this time, the user can trigger model retraining to let the model learn new operating data to adapt to the new working conditions. The user can also manually adjust the control parameters of the air-conditioning equipment directly. For example, under certain special production process requirements, the temperature and humidity of the workshop need to be changed quickly, and the automatic control strategy cannot respond in time. The user can manually adjust the parameters such as the opening of the fresh air valve and the frequency of the blower to meet the urgent needs of production.

[0077] In addition, when users find that the process air conditioning is operating abnormally or are dissatisfied with the control effect, they can manually trigger the self-optimization of the model to retrain and optimize the model to meet actual production needs.

[0078] In the embodiment of the present invention, the triggering condition includes: reaching a preset evaluation time and / or the number of responses to the intervention instruction within a preset time period exceeds a preset value.

[0079] In some operating systems, due to the limited computing power of computers, it is not necessary to use the GRU model to evaluate every time the MLP model generates control parameters. The evaluation trigger conditions can be set to reduce the number of GRU model evaluations. For example, set an evaluation every 12 hours, or if the number of intervention instructions appears too many times within a period of time, indicating that the user is not satisfied with the actual operation effect, then use the GRU model to evaluate the control parameters, and decide whether to perform self-optimization of the MLP model based on the evaluation results.

[0080] In addition, the optimal control parameters calculated by the control model are transmitted to the OPC (Open Platform Communication) interface in real time to ensure that the control instructions can be accurately sent to the PLC (Programmable Logic Controller) of the process air conditioner. The control parameters are encoded in a predetermined format to ensure the compatibility and real-time communication with the equipment, thereby achieving precise adjustment and control of the air-conditioning equipment. Through the OPC interface, the output module can ensure the efficient and stable transmission of control parameters between the equipment and the system, and ensure the stable operation and optimized adjustment of the air-conditioning system. The PLC execution module receives the control parameters from the output module and converts them into specific control instructions, which directly act on the hardware components of the air-conditioning equipment.

[0081] The solution of the embodiment of the present invention has at least the following advantages and improvements:

[0082] In order to solve the problem of insufficient intelligence and adaptability, a multi-layer perceptron model is used to dynamically learn and predict the air conditioner operation data and automatically adjust the control parameters. This design enables the system to adapt to environmental changes, the degradation of sensing devices and execution devices, avoids the limitation of traditional industrial control systems that can only operate according to fixed logic programs, and improves the intelligent control level of air conditioners.

[0083] In order to solve the problem of insufficient data processing and analysis capabilities, the ETL technology is used to extract structured and unstructured data such as relational data and flat data files from various scattered and heterogeneous data sources generated in the production process to a temporary intermediate layer for cleaning and conversion. The processed data is analyzed in real time by combining the mechanism and big data analysis methods to fully tap the potential value and provide a basis for the continuous optimization of the control system. This design makes up for the deficiencies of traditional industrial control systems in data processing and analysis capabilities.

[0084] In order to solve the problem of insufficient optimization of energy utilization, the control model (trained MLP model) outputs the air conditioning control plan in real time, and the evaluation algorithm (GRU model) and the manual judgment module jointly decide the self-optimization of the air conditioning control model and whether the plan should be issued. According to the real-time control plan, the working state of the equipment can be automatically adjusted to achieve the best energy efficiency utilization, reduce energy waste, and solve the problem of low energy efficiency of traditional industrial control systems.

[0085] Embodiment 2

[0086] Figure 2 Schematic diagram of the structure of a process air conditioning control device based on a prediction model provided in the second embodiment of the present invention. Figure 2 As shown, the device comprises:

[0087] The data collection unit 210 is used to collect real-time data of the operation site where the process air conditioner is located, and pre-process the real-time data to generate standardized time series data;

[0088] The parameter generation unit 220 is used to input the standardized time series data into the multi-layer perceptron MLP model to generate the process air conditioning control parameters under the current working conditions;

[0089] The performance evaluation unit 230 is used to perform performance evaluation on the control parameters through the gated recurrent unit GRU model when the preset evaluation trigger conditions are met. When the evaluation results meet the expected control effect, the control parameters are transmitted to the PLC of the process air conditioner to adjust the control parameters of the process air conditioner.

[0090] Optionally, the data acquisition unit 210 is specifically configured to execute:

[0091] Collect real-time data from the air conditioning system, boiler system, refrigeration system and MES system at the work site. The real-time data includes the opening of the fresh air valve, the opening of the return air valve, the opening of the chilled water valve, the opening of the heating valve, the opening of the humidification valve, the opening of the supply air valve, the frequency of the supply fan, the temperature and pressure of the chilled water, the temperature and pressure of the heating steam, the temperature and pressure of the humidification steam, the temperature and humidity of the fresh air, the temperature and humidity of the return air, the temperature and humidity of the supply air, the temperature and humidity of the sensor in the mixing workshop, and the opening of the air duct in the mixing workshop;

[0092] Process missing values ​​and outliers in real-time data and perform data standardization or normalization;

[0093] Data features are extracted based on correlation coefficients, moving averages, and time trends to generate standardized time series data.

[0094] Optionally, the parameter generating unit 220 is specifically configured to execute:

[0095] The historical data of process air conditioning is used for model training, the output of process air conditioning is predicted, and the optimization target is solved to generate control strategies.

[0096] Optionally, the performance evaluation unit 230, when performing the performance evaluation of the control parameters through the gated recurrent unit GRU model, specifically performs:

[0097] Receive the control parameters output from the MLP model and the historical performance data of the process air conditioner, and conduct performance evaluation in terms of response speed, energy saving effect and environmental adaptability. The evaluation indicators are the real-time response time of the air conditioning system, the temperature and humidity control accuracy and the energy consumption optimization ratio;

[0098] Capture the dynamic changes and dependencies of input data based on recurrent neural networks;

[0099] As for the dynamic changes and dependencies, the real-time response time, temperature and humidity control accuracy, and energy consumption optimization ratio are quantitatively evaluated.

[0100] Optionally, the performance evaluation unit 230, when performing the performance evaluation of the control parameters through the gated recurrent unit GRU model, specifically performs:

[0101] Analyze the real-time operation data and historical feedback data of process air conditioning and calculate the deviation between actual performance and target effect;

[0102] When the deviation is less than the preset range, it is determined that the evaluation result meets the expected control effect; if the deviation is not less than the preset range, the MLP model is self-optimized, and the self-optimization includes adjusting the parameters and training strategy of the MLP model according to the deviation and updating the model structure of the MLP model.

[0103] Optionally, the device may further include: an intervention execution module;

[0104] An intervention execution module, used to respond to an external intervention instruction, wherein the intervention instruction includes at least one control parameter of the process air conditioner;

[0105] Adjust the process air conditioning control parameters under the current working conditions according to the control parameters.

[0106] Optionally, the triggering conditions include: reaching a preset evaluation time and / or the number of responses to the intervention instruction within a preset time period exceeds a preset value.

[0107] The process air conditioning control device based on the prediction model provided in the embodiment of the present invention can execute the process air conditioning control method based on the prediction model provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0108] Embodiment 3

[0109] Figure 3 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can 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.

[0110] 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. An input / output (I / O) interface 15 is also connected to the bus 14.

[0111] 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.

[0112] 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 a process air conditioning control method based on a prediction model.

[0113] In some embodiments, the process air conditioning control method based on the prediction model can 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 can 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 process air conditioning control method based on the prediction model described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the process air conditioning control method based on the prediction model in any other appropriate manner (for example, by means of firmware).

[0114] 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.

[0115] 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.

[0116] 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.

[0117] 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).

[0118] 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.

[0119] 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.

[0120] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0121] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A process air conditioning control method based on a prediction model, characterized in that: include: Real-time data of the operation site where the process air conditioner is located is collected in real time, and the real-time data is preprocessed to generate standardized time series data; Inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions; When the preset evaluation trigger conditions are met, the control parameters are performance evaluated through the gated recurrent unit GRU model. When the evaluation results meet the expected control effects, the control parameters are transmitted to the PLC of the process air conditioner to adjust the control parameters of the process air conditioner.

2. The method according to claim 1, characterized in that The real-time collection of multi-source heterogeneous data at the work site where the process air conditioner is located, and pre-processing the multi-source heterogeneous data to generate standardized time series data, include: Collect the real-time data from the air conditioning system, boiler system, refrigeration system and MES system of the work site, the real-time data including fresh air valve opening, return air valve opening, chilled water valve opening, heating valve opening, humidification valve opening, supply air valve opening, supply fan frequency, chilled water temperature and pressure, heating steam temperature and pressure, humidification steam temperature and pressure, fresh air temperature and humidity, return air temperature and humidity, supply air temperature and humidity, blending workshop sensor temperature and humidity and blending workshop air duct opening; Processing missing values ​​and outliers in the real-time data and performing data standardization or normalization; Data features are extracted based on correlation coefficient, moving average and time trend to generate the standardized time series data.

3. The method according to claim 1, characterized in that The step of inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions includes: The historical data of process air conditioning is used for model training, the output of process air conditioning is predicted, and the optimization target is solved to generate control strategies.

4. The method according to claim 1, characterized in that The performance evaluation of the control parameters by using a gated recurrent unit (GRU) model includes: Receiving control parameters outputted from the MLP model and historical performance data of the process air conditioner, and conducting performance evaluation in terms of response speed, energy-saving effect and environmental adaptability, wherein the evaluation indicators are the real-time response time of the air conditioning system, the temperature and humidity control accuracy and the energy consumption optimization ratio; Capture the dynamic changes and dependencies of input data based on recurrent neural networks; As for the dynamic changes and dependencies, the real-time response time, temperature and humidity control accuracy, and energy consumption optimization ratio are quantitatively evaluated.

5. The method according to claim 1, characterized in that: The performance evaluation of the control parameters by using a gated recurrent unit (GRU) model includes: Analyze the real-time operation data and historical feedback data of the process air conditioner to calculate the deviation between the actual performance and the target effect; When the deviation is less than a preset range, it is determined that the evaluation result meets the expected control effect; if the deviation is not less than the preset range, the MLP model is self-optimized, and the self-optimization includes adjusting the parameters and training strategy of the MLP model according to the deviation and updating the model structure of the MLP model.

6. The method according to claim 1, characterized in that Further including: responding to an external intervention instruction, wherein the intervention instruction includes at least one control parameter of the process air conditioner; The process air conditioning control parameters under the current working conditions are adjusted according to the control parameters.

7. The method according to claim 6, characterized in that The triggering condition includes: reaching a preset evaluation time and / or the number of responses to the intervention instruction within a preset time period exceeds a preset value.

8. A process air conditioning control device based on a prediction model, characterized in that: include: A data acquisition unit, used for real-time acquisition of real-time data of the operation site where the process air conditioner is located, and preprocessing the real-time data to generate standardized time series data; A parameter generation unit, used for inputting the standardized time series data into a multi-layer perceptron MLP model to generate process air conditioning control parameters under current working conditions; The performance evaluation unit is used to perform performance evaluation on the control parameters through a gated recurrent unit (GRU) model when a preset evaluation trigger condition is met, and when the evaluation result meets the expected control effect, the control parameters are transmitted to the PLC of the process air conditioner to adjust the control parameters of the process air conditioner.

9. An electronic device, characterized in that: The electronic device comprises: At least one processor; and a memory communicatively connected to the at least one processor; wherein 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 process air conditioning control method based on a predictive model as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the process air conditioning control method based on a prediction model as described in any one of claims 1 to 7 when executed.

11. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the process air conditioning control method based on a prediction model according to any one of claims 1 to 7.

Images