Chemical fiber temperature and humidity monitoring method and system based on LSTM neural network

Abstract

The invention discloses a chemical fiber temperature and humidity monitoring method and system based on an LSTM neural network. Temperature data and humidity data in a monitoring area are collected through a distributed temperature sensor and a distributed humidity sensor; the temperature data and the humidity data are input into a temperature and humidity optimization neural network model, a control instruction for execution equipment is output, and the execution equipment comprises an air conditioner and a humidifier; the execution device adjusts the temperature and humidity of the monitoring area according to the control instruction, the air conditioner and the humidifier execute accurate regulation and control, accurate control over the temperature and humidity of the chemical fiber storage area is achieved, and chemical fiber storage safety is protected.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical fiber warehousing monitoring, and in particular to a method and system for monitoring the temperature and humidity of chemical fibers based on an LSTM neural network. Background Art

[0002] During the storage of chemical fibers, the stability of temperature and humidity parameters directly affects the storage safety. Traditional monitoring methods mostly use fixed threshold to control air conditioners and humidifiers, and have the following defects: they cannot dynamically adjust the temperature and humidity target values according to the production stage; it is easy to generate energy consumption conflicts when the air conditioner and the humidifier are controlled in coordination; the response to the different temperature and humidity requirements of multiple regions lags behind.

[0003] In view of the above situation, it is urgent for the staff in this field to provide a monitoring solution for the temperature and humidity of chemical fibers with self-learning ability. Summary of the Invention

[0004] The present invention provides a method and system for monitoring the temperature and humidity of chemical fibers based on an LSTM neural network, so as to solve the technical problem of the lack of a monitoring solution for the temperature and humidity of chemical fibers with self-learning ability in the prior art.

[0005] To solve the above technical problem, the technical solution proposed by the present invention is: a method for monitoring the temperature and humidity of chemical fibers based on an LSTM neural network, including:

[0006] Collecting temperature data and humidity data in the monitoring area through distributed temperature sensors and distributed humidity sensors;

[0007] Inputting the temperature data and humidity data into a temperature and humidity optimization neural network model, and outputting control instructions for the execution devices, where the execution devices include air conditioners and humidifiers;

[0008] The execution devices adjust the temperature and humidity in the monitoring area according to the control instructions.

[0009] Optionally, the training method of the temperature and humidity optimization neural network model includes the following:

[0010] Experts mark the optimal temperature and the optimal humidity;

[0011] Collecting historical parameters in the monitoring area within a set time, including temperature, humidity and the corresponding control instructions for air conditioners and humidifiers, to form a sample set;

[0012] Dividing the sample set into a training set, a validation set and a test set according to a set ratio;

[0013] Initializing the temperature and humidity optimization neural network model, training the temperature and humidity optimization neural network model with the training set, and monitoring overfitting through the validation set;

[0014] Set a loss function and perform iterative training on the temperature and humidity optimization neural network model through backpropagation and optimization methods. When the set number of iteration rounds is reached or the loss function value no longer decreases after the set number of iteration rounds, stop the iterative training to obtain a converged temperature and humidity optimization neural network model.

[0015] Optionally, the following loss function is used for training: L = α·(T pred - T true ) 2 + β·(H pred - H true ) 2 , where T pred is the temperature control value predicted by the model, T true is the actual required temperature adjustment amount, H pred is the humidity control value predicted by the model, H true is the actual required humidity adjustment amount, α is the temperature weight coefficient, β is the humidity weight coefficient, and α + β = 1.

[0016] Optionally, it further includes: using a test set to test the temperature and humidity optimization neural network model, calculating the error. If the error reaches the set value, output the temperature and humidity optimization neural network model; if the error does not reach the set value, continue to perform iterative training on the temperature and humidity optimization neural network model until the error reaches the set value.

[0017] Optionally, after the control instruction is generated, the temperature and humidity data after the adjustment of the execution device are collected in real time and fed back to the neural network model for closed-loop control to dynamically correct the model output.

[0018] The present invention also provides a chemical fiber temperature and humidity monitoring system based on an LSTM neural network, including a distributed temperature sensor, a distributed humidity sensor, a regulator, and an execution device. The distributed temperature sensor and the distributed humidity sensor are arranged in the monitoring area. The distributed temperature sensor and the distributed humidity sensor are respectively used to collect the temperature data and humidity data in the monitoring area and transmit the temperature data and humidity data to the regulator;

[0019] The regulator is used to receive the temperature data and humidity data and output a control instruction for the execution device;

[0020] The execution device adjusts the temperature and humidity of the monitoring area according to the control instruction.

[0021] The present invention has the following beneficial effects: The method of the present invention collects temperature data and humidity data in the monitoring area through a distributed temperature sensor and a distributed humidity sensor; inputs the temperature data and humidity data into a temperature and humidity optimization neural network model, and outputs control instructions for the execution devices, where the execution devices include an air conditioner and a humidifier; the execution devices adjust the temperature and humidity in the monitoring area according to the control instructions, and the air conditioner and the humidifier perform precise regulation, achieving precise control of the temperature and humidity in the chemical fiber storage area and protecting the safety of chemical fiber storage.

[0022] In addition to the objectives, features, and advantages described above, the present invention has other objectives, features, and advantages. The following will refer to the accompanying drawings for a more detailed description of the present invention. Description of the Drawings

[0023] The drawings forming a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0024] Figure 1 is a flowchart of a chemical fiber temperature and humidity monitoring method based on an LSTM neural network according to a preferred embodiment of the present invention; Detailed Embodiments

[0025] The following will describe the embodiments of the present invention in detail with reference to the accompanying drawings. However, the present invention can be implemented in many different ways defined and covered by the claims.

[0026] See Figure 1 , a chemical fiber temperature and humidity monitoring method based on an LSTM neural network of the present invention includes:

[0027] S101 Collect temperature data and humidity data in the monitoring area through a distributed temperature sensor and a distributed humidity sensor;

[0028] S102 Input the temperature data and humidity data into a temperature and humidity optimization neural network model, and output control instructions for the execution devices, where the execution devices include an air conditioner and a humidifier;

[0029] S103 The execution devices adjust the temperature and humidity in the monitoring area according to the control instructions.

[0030] Optionally, the training method of the temperature and humidity optimization neural network model includes the following:

[0031] Experts mark the optimal temperature and the optimal humidity;

[0032] Collect historical parameters in the monitoring area within a set time, including temperature, humidity, and corresponding control instructions for the air conditioner and the humidifier, to form a sample set;

[0033] Divide the sample set into a training set, a validation set, and a test set according to a set ratio; the ratio of the training set, the validation set, and the test set is 7:2:1.

[0034] Initialize the temperature and humidity optimization neural network model, train the temperature and humidity optimization neural network model using the training set, and monitor overfitting through the validation set;

[0035] Set the loss function, and perform iterative training on the temperature and humidity optimization neural network model through backpropagation and optimization methods. When the set number of iteration rounds is reached or the loss function value no longer decreases after a set number of iteration rounds, stop the iterative training to obtain a converged temperature and humidity optimization neural network model.

[0036] Optionally, the following loss function is used for training: L = α·(T pred -T true ) 2 +β·(H pred -H true ) 2 , where T pred is the temperature control value predicted by the model, T true is the actual required temperature adjustment amount, H pred is the humidity control value predicted by the model, H true is the actual required humidity adjustment amount, α is the temperature weight coefficient, β is the humidity weight coefficient, and α + β = 1.

[0037] In this embodiment, α = 0.3 and β = 0.7.

[0038] Optionally, it further includes: using the test set to test the temperature and humidity optimization neural network model, calculating the error. If the error reaches the set value, output the temperature and humidity optimization neural network model; if the error does not reach the set value, continue to perform iterative training on the temperature and humidity optimization neural network model until the error reaches the set value.

[0039] Optionally, after the control instruction is generated, the temperature and humidity data adjusted by the execution device are collected in real time and fed back to the neural network model for closed-loop control to dynamically correct the model output.

[0040] Example 2:

[0041] A chemical fiber temperature and humidity monitoring system based on an LSTM neural network, including a distributed temperature sensor, a distributed humidity sensor, a regulator, and an execution device. The distributed temperature sensor and the distributed humidity sensor are arranged in the monitoring area. The distributed temperature sensor and the distributed humidity sensor are respectively used to collect the temperature data and humidity data in the monitoring area and transmit the temperature data and humidity data to the regulator;

[0042] The system can implement each embodiment of the above-mentioned water level prediction method based on the NSGA-III optimized LSTM neural network and can achieve the same beneficial effects, which will not be elaborated here.

[0043] In summary, the present invention obtains the water level data before the time point to be predicted, preprocesses the water level data, inputs the preprocessed water level data into the water level prediction model, and outputs the water level prediction result. The water level prediction model is based on the LSTM neural network framework and uses NSGA-III for parameter optimization. The water level prediction model of the present invention has stronger stability, and considers the multi-objective parameter optimization process. On the premise of ensuring good prediction performance, it can effectively reduce the model calculation amount. Using the water level prediction model of the present invention can accurately and effectively predict the water level, overcoming the disadvantages of complex solution and poor transferability of the traditional water level prediction model.

[0044] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0045] Any process or method description represented in the flowchart or described in other ways herein can be understood as representing a module, segment or part of code including one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the technical field of the embodiments of the present invention.

[0046] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processing module, or other systems that can fetch and execute instructions from the instruction execution system, apparatus or device), or in combination with these instruction execution systems, apparatus or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transport a program for use by or in connection with an instruction execution system, apparatus or device.

[0047] It should be understood that each part of the embodiments of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following technologies well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0048] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for monitoring temperature and humidity of chemical fibers based on LSTM neural network, characterized in that: include: Collect temperature data and humidity data in the monitoring area through distributed temperature sensors and distributed humidity sensors; Input the temperature data and the humidity data into a temperature and humidity optimization neural network model, and output a control instruction for an execution device, wherein the execution device includes an air conditioner and a humidifier; The execution device adjusts the temperature and humidity of the monitored area according to the control instruction.

2. The method for monitoring temperature and humidity of chemical fibers based on LSTM neural network according to claim 1, characterized in that: The training method of the temperature and humidity optimization neural network model includes the following: Experts mark the optimal temperature and optimal humidity; Collect historical parameters of the monitoring area within the set time, including temperature, humidity and corresponding air conditioning and humidifier control instructions, to form a sample set; Divide the sample set into training set, validation set and test set according to the set ratio; Initializing the temperature and humidity optimization neural network model, training the temperature and humidity optimization neural network model using a training set, and monitoring overfitting through a validation set; A loss function is set, and the temperature and humidity optimization neural network model is iteratively trained through back propagation and optimization methods. When the set number of iterations is reached or the loss function value no longer decreases after the set number of iterations, the iterative training is stopped to obtain a converged temperature and humidity optimization neural network model.

3. The method for monitoring temperature and humidity of chemical fiber based on LSTM neural network according to claim 2, characterized in that: The following loss function is used for training: L = α·(T pred -T true ) 2 +β·(H pred -H true ) 2 , where T pred is the temperature control value predicted by the model, T true is the actual required temperature adjustment, H pred is the humidity control value predicted by the model, H true is the actually required humidity adjustment amount, α is the temperature weight coefficient, β is the humidity weight coefficient, and α+β=1.

4. The method for monitoring temperature and humidity of chemical fibers based on LSTM neural network according to claim 3, characterized in that: It is characterized in that It also includes: using the test set to test the temperature and humidity optimization neural network model, calculating the error, and if the error reaches a set value, outputting the temperature and humidity optimization neural network model; if the error does not reach the set value, continuing to iteratively train the temperature and humidity optimization neural network model until the difference reaches the set value.

5. The method for monitoring temperature and humidity of chemical fiber based on LSTM neural network according to claim 2, characterized in that: After the control instruction is generated, the temperature and humidity data after the execution equipment adjustment is collected in real time and fed back to the neural network model for closed-loop control to dynamically correct the model output.

6. A chemical fiber temperature and humidity monitoring system based on LSTM neural network, characterized in that: It includes a distributed temperature sensor, a distributed humidity sensor, a regulator and an execution device, wherein the distributed temperature sensor and the distributed humidity sensor are arranged in a monitoring area, and the distributed temperature sensor and the distributed humidity sensor are used to collect temperature data and humidity data in the monitoring area respectively, and transmit the temperature data and humidity data to the regulator; The regulator is used to receive temperature data and humidity data and output control instructions for the execution device; The execution device adjusts the temperature and humidity of the monitoring area according to the control instruction.

Images