A tunnel fan intelligent control method and system based on smoke prediction

Abstract

The present invention belongs to the technical field of ventilation control. Provided is a method and system for intelligent control of a tunnel fan based on smoke prediction. The method comprises: collecting and using training data to train a BiLSTM model; a sensor data acquisition module collects environmental data in the tunnel, obtains first smoke detector data, first temperature sensor data and first humidity sensor data, and simultaneously inputs the environmental data into the trained BiLSTM model to obtain predicted second smoke detector data, second temperature sensor data and second humidity sensor data at a future moment; compares the second smoke detector data with the first interval, compares the second temperature sensor data with the second interval, and compares the second humidity sensor data with the third interval, and intelligently controls the tunnel fan according to the comparison result. The present invention can automatically adjust the fan based on the prediction of a fire, and can timely discharge smoke from the tunnel.

Description

Technical Field

[0001] The present invention relates to the technical field of ventilation control, and in particular to a method and system for intelligently controlling a tunnel fan based on smoke prediction. Background Art

[0002] Tunnel fans are key equipment in tunnel ventilation systems. In addition to ensuring air circulation in tunnels, removing harmful gases and maintaining driving safety, they can also be used to control smoke diffusion in tunnels. However, existing tunnel fans work according to preset ventilation parameters when controlling smoke diffusion in tunnels, and cannot adapt to the actual situation of smoke in tunnels, resulting in poor control of smoke diffusion in tunnels. The present invention aims to improve this technical problem. Summary of the invention

[0003] In response to the above technical problems, the present invention provides a method, system, electronic device, computer storage medium and computer program product for intelligent control of tunnel fans based on smoke prediction to improve the efficiency and ease of use of intelligent control of tunnel fans based on smoke prediction.

[0004] The invention discloses an intelligent control method for a tunnel fan based on smoke prediction, the method comprising the following steps: collecting and using training data to train a BiLSTM model; a sensor data acquisition module acquires environmental data in a tunnel to obtain first smoke detector data, first temperature sensor data and first humidity sensor data, the first smoke detector data, the first temperature sensor data and the first humidity sensor data are simultaneously input into the trained BiLSTM model, the BiLSTM model outputs predicted second smoke detector data, second temperature sensor data and second humidity sensor data at a future moment; the second smoke detector data is compared with a first interval, the second temperature sensor data is compared with a second interval, the second humidity sensor data is compared with a third interval, and the tunnel fan is intelligently controlled according to the comparison result; wherein the intelligent control comprises intelligent start, intelligent speed regulation and intelligent stop.

[0005] Optionally, the collecting and using training data to train the BiLSTM model includes: a sensor data acquisition module collecting environmental data in the tunnel, the environmental data including a third smoke detector data, a third temperature sensor data and a third humidity sensor data; based on the environmental data, a model initialization step, a training set generation step and a training process step are respectively performed to realize the training of the BiLSTM model.

[0006] Optionally, the intelligent control of the tunnel fan according to the comparison result includes: if the second smoke detector data reaches , then the tunnel fan is controlled to start running; if the second smoke detector data reaches , then the tunnel fan is controlled to run at full speed, that is, the smoke detector data is in the range When the second temperature sensor data reaches , then the tunnel fan is controlled to start running; if the second temperature sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the temperature sensor data is in the range When the tunnel fan is controlled to be in the running stage; if the second humidity sensor data reaches , then the tunnel fan is controlled to start running; if the second humidity sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the humidity sensor data is in the range When the tunnel fan is controlled, it is in the operation stage.

[0007] Optionally, the intelligent control of the tunnel fan according to the comparison result further includes: setting the weight of the smoke detector value to , set the weight of the temperature sensor value to , set the weight of the humidity sensor value to ; Based on the first interval , the second interval and the third interval Calculate the comprehensive interval , calculated as follows: ; Calculate the mean of each value in the second smoke detector data, the second temperature sensor data, and the second humidity sensor data respectively, that is, obtain ;right and Calculate and operate according to the following formula: ;if , then stop the tunnel fan; if , then let the tunnel fan run at full speed.

[0008] Optionally, the intelligent control of the tunnel fan according to the comparison result further includes: if , then calculate the power of the tunnel fan to be operated according to the following formula: ; In the formula, is the power of the tunnel fan running at full speed; The value of is used to control the speed of the tunnel fan using the PID control algorithm.

[0009] Optionally, the intelligent control of the tunnel fan according to the comparison result further includes: performing equivalent calculation on the number of people and vehicles in the tunnel to obtain the equivalent number of people and vehicles, determining a reinforcement coefficient based on the equivalent number of people and vehicles, and using the reinforcement coefficient to adjust the power Make corrections and control the tunnel fan according to the corrected power run.

[0010] The present invention also discloses an intelligent control system for a tunnel fan based on smoke prediction, the system comprising a processing device and a storage device, the computer code stored in the storage device being called and executed by the processing device to implement the following steps: collecting and using training data to train a BiLSTM model; a sensor data acquisition module acquiring environmental data in the tunnel to obtain first smoke detector data, first temperature sensor data and first humidity sensor data, inputting the first smoke detector data, the first temperature sensor data and the first humidity sensor data into the trained BiLSTM model at the same time, the BiLSTM model outputting predicted second smoke detector data, second temperature sensor data and second humidity sensor data at a future moment; comparing the second smoke detector data with the first interval, comparing the second temperature sensor data with the second interval, comparing the second humidity sensor data with the third interval, and intelligently controlling the tunnel fan according to the comparison results; wherein the intelligent control includes intelligent start, intelligent speed regulation and intelligent stop.

[0011] The present invention also discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement any of the above methods.

[0012] The present invention also discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any of the above methods.

[0013] The present invention also discloses a computer program product, which includes computer codes. When the computer codes are executed by a processor of an electronic device, any of the above methods is implemented.

[0014] The solution of the present invention does not require human intervention and can achieve intelligent start, intelligent speed regulation, and intelligent stop, thereby reducing the operational burden of operation and maintenance personnel; in emergency situations, such as a fire in a tunnel, the system can automatically and timely adjust the fan speed based on the prediction of the disaster, without the need for operation and maintenance personnel to manually adjust the fan speed after being aware of the fire on site, so as to discharge smoke from the tunnel in time and increase the escape time of trapped people. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a flow chart of a tunnel fan intelligent control method based on smoke prediction disclosed in an embodiment of the present invention.

[0017] Figure 2 It is a structural schematic diagram of a tunnel fan intelligent control system based on smoke prediction disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0019] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0020] like Figure 1 As shown, an embodiment of the present invention discloses an intelligent control method for a tunnel fan based on smoke prediction, the method comprising the following steps: collecting and using training data to train a BiLSTM model; a sensor data acquisition module acquires environmental data in the tunnel to obtain first smoke detector data, first temperature sensor data and first humidity sensor data, and simultaneously inputs the first smoke detector data, the first temperature sensor data and the first humidity sensor data into the trained BiLSTM model, and the BiLSTM model outputs the predicted second smoke detector data, second temperature sensor data and second humidity sensor data at a future moment; the second smoke detector data is compared with the first interval, the second temperature sensor data is compared with the second interval, the second humidity sensor data is compared with the third interval, and the tunnel fan is intelligently controlled according to the comparison results; wherein the intelligent control includes intelligent start, intelligent speed regulation and intelligent stop.

[0021] The present invention uses the BiLSTM model to perform in-depth analysis on the collected environmental data in the tunnel, so as to predict the environmental data in the tunnel at future moments, and then by comparing the environmental data at future moments with a preset numerical range, the control mode of the tunnel fan can be determined, thereby realizing intelligent control of the tunnel fan.

[0022] The solution of the present invention does not require human intervention and can achieve intelligent start, intelligent speed regulation, and intelligent stop, thereby reducing the operational burden of operation and maintenance personnel; in emergency situations, such as a fire in a tunnel, the system can automatically and timely adjust the fan speed based on the prediction of the disaster, without the need for operation and maintenance personnel to manually adjust the fan speed after being aware of the fire on site, so as to discharge smoke from the tunnel in time and increase the escape time of trapped people.

[0023] Optionally, the collecting and using training data to train the BiLSTM model includes: a sensor data acquisition module collecting environmental data in the tunnel, the environmental data including a third smoke detector data, a third temperature sensor data and a third humidity sensor data; based on the environmental data, a model initialization step, a training set generation step and a training process step are respectively performed to realize the training of the BiLSTM model.

[0024] The specific training process is exemplified as follows: Step 1-1: The sensor data acquisition module acquires environmental data in the tunnel, where the environmental data includes third smoke detector data, third temperature sensor data, and third humidity sensor data.

[0025] For example, smoke detectors, temperature sensors, and humidity sensors installed in tunnels collect data at a frequency of once per second. The data collected by smoke detectors is used express, Indicates The smoke detector values ​​collected at all times; the temperature sensor data collected is used express, Indicates The temperature sensor values ​​collected at all times; the humidity sensor data collected is used express, Indicates The humidity sensor values ​​collected at all times, .

[0026] Step 1-2: Set a set of trainable weight parameters, ; In the formula, and The matrix size of each element in is .

[0027] Set a set of trainable bias parameters, ; In the formula, The matrix size of each element in is .

[0028] Setting trainable weight parameters , The matrix size is ; Set trainable bias parameters , The matrix size is .

[0029] Step 1-3: Set the hidden layer size of the BiLSTM model to 600 and initialize all the weight parameters of the BiLSTM model and all bias parameters .

[0030] The above steps 1-2 and 1-3 are model initialization steps.

[0031] Step 1-4: Set up list variables , and set it to an empty list as the training set input data list; set the list variable , and let it be an empty list as the sample label list of the training set.

[0032] Step 1-5: Set variables And order .

[0033] Step 1-6: Set variables And order ; If the list has been set , then set it to an empty list, if no list has been set , then set the list , and let it be an empty list.

[0034] Step 1-7: Follow the following formula and method: ;at this time The size is ,Will Perform horizontal splicing into tensors, ;at this time The size is , and Add to list Tail.

[0035] Step 1-8: If , then let the variable Increment by 1, and then repeat steps 1-7 to 1-8; otherwise, execute steps 1-9.

[0036] Step 1-9: List Store the value of at the end of the list variable Reset to an empty list.

[0037] Step 1-10: Set up the list ,Will Put values ​​into list variables The tail, then Put into list variable The tail, then Put into list variable Tail.

[0038] Step 1-11: List Store the value of at the end of the list variable Reset to an empty list.

[0039] Step 1-12: If , then let the variable Increment by 1, and then repeat steps 1-6 to 1-11; otherwise, execute step 1-13.

[0040] Step 1-13: So far, the length is List ,and Each list in is 600 long; and the length is List ,and Each list in is of length 3.

[0041] The above steps 1-4 to 1-13 are steps for generating a training set.

[0042] Step 1-14: Generate a tolerance of 1 from 1 to Numbers, using list variables Store them in ascending order, that is .

[0043] Step 1-15: Randomly shuffle the list variables The order of the numbers in .

[0044] Step 1-16: Set variables And set its value to 1.

[0045] Step 1-17: Get the list variable From the first To A total of 32 values, and set a list vector , store these values ​​into a list vector , then the list variable The length of is 32.

[0046] Step 1-18: Get the list No. There are 32 elements in total, set the list variable And store these elements in the list variable in the order they are taken out ;in, Refers to the list Middle elements.

[0047] Step 1-19: List Input to the BiLSTM model set in step 1-13 and get the output Tensor, that is ;in, Size .

[0048] Step 1-20: Follow the following formula and method: ;in, The size is .

[0049] Step 1-21: Get the list No. There are 32 elements in total, set the list variable And store these elements in the list variable in the order they are taken out ,at this time The length of is 32, where each element is of size 3. Therefore, Considered as size Tensor of .

[0050] Step 1-22: Convert the tensor The dimension is compressed to , that is, The dimensions are consistent.

[0051] Step 1-23: Calculate and operate according to the following formula: ;in, Refers to the tensor The order in the first dimension is The value of , the order in the second dimension is The numerical value of .

[0052] Step 1-24: Use the following formula and operation Calculate the gradient, ;in, Refers to finding partial derivatives of all weight parameters in the BiLSTM model. Refers to taking partial derivatives of all bias parameters in the BiLSTM model.

[0053] Step 1-25: Based on the results calculated in step 1-24, update the values ​​of each weight parameter and bias parameter according to the following formula and operation: ;in, is the learning rate (can be changed according to actual conditions, generally set to 0.002).

[0054] Step 1-26: If The value is greater than , then execute steps 1-27, otherwise, let the variable Increment the value by 1 and execute steps 1-17 to 1-26.

[0055] The above steps 1-14 to 1-26 are training process steps.

[0056] Step 1-27: At this point, the training system is completed. The BiLSTM model parameters are saved for subsequent execution of the system.

[0057] In addition, the specific analysis and processing process of the BiLSTM model is exemplified as follows: at any time, environmental data in the tunnel is collected, including second smoke detector data, second temperature sensor data, and second humidity sensor data.

[0058] The second smoke detector, the second temperature sensor, and the second humidity sensor all collect data at a frequency of once per second, collecting a total of 600 seconds of data. The collected data from the smoke detector is used express, Indicates The smoke detector values ​​collected at all times; the temperature sensor data collected is used express, Indicates The temperature sensor values ​​collected at all times; the humidity sensor data collected is used express, Indicates The humidity sensor value collected at all times.

[0059] Step 2-1: Load the BiLSTM model trained by the training system and And other parameters.

[0060] Step 2-2: Set the list variable , and let it be an empty list as the training set input data list.

[0061] Step 2-3: Set variables And order .

[0062] Step 2-4: Set variables And order .

[0063] Step 2-5: Operate according to the following formula and method. ;at this time The size is ,Will Perform horizontal splicing into tensors, ,at this time The size is , and Add to list Tail.

[0064] Step 2-6: If , then let the variable Increment by 1, and then repeat steps 2-5 to 2-6. Otherwise, execute step 2-7.

[0065] Step 2-7: List Input to the BiLSTM model loaded in step 2-1 and get the output Tensor, that is ,in, Size .

[0066] Step 2-8: Follow the following formula and method: ; Among them, the tensor The size is .

[0067] Step 2-9: Convert the tensor The dimension of is compressed to a tensor of size 3. So far, we can see that the tensor The first element refers to the model's value for the smoke. Time prediction, using To represent. The second element refers to the model's temperature value. Time prediction, using To represent. The third element refers to the model's calculation of the humidity value. Time prediction, using To express.

[0068] Step 2-10: If , then let the variable Increment by 1, and then repeat steps 2-5 to 2-9. Otherwise, execute step 2-11.

[0069] Step 2-11: So far, the model has completed the prediction of the second smoke detector data, the second temperature sensor data, and the second humidity sensor data for 180 moments from the 601st moment (inclusive) to the 780th moment (inclusive), that is, the next 180 seconds from the current moment.

[0070] Optionally, the intelligent control of the tunnel fan according to the comparison result includes: if the second smoke detector data reaches , then the tunnel fan is controlled to start running; if the second smoke detector data reaches , then the tunnel fan is controlled to run at full speed, that is, the smoke detector data is in the range When , the tunnel fan is controlled to be in the operation stage.

[0071] If the second temperature sensor data reaches , then the tunnel fan is controlled to start running; if the second temperature sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the temperature sensor data is in the range When the tunnel fan is controlled, it is in the operation stage.

[0072] If the second humidity sensor data reaches , then the tunnel fan is controlled to start running; if the second humidity sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the humidity sensor data is in the range When the tunnel fan is controlled, it is in the operation stage.

[0073] In this embodiment, by comparing the three environmental data collected above with the preset value intervals, namely the first interval, the second interval, and the third interval, the specific control strategy to be adopted for the tunnel fan can be determined. , the second interval is , the third interval is .

[0074] Optionally, the intelligent control of the tunnel fan according to the comparison result further includes: setting the weight of the smoke detector value to , set the weight of the temperature sensor value to , set the weight of the humidity sensor value to .

[0075] Based on the first interval , the second interval and the third interval Calculate the comprehensive interval , calculated as follows: ; Calculate the mean of each value in the second smoke detector data, the second temperature sensor data, and the second humidity sensor data respectively, that is, obtain .

[0076] right Calculate and operate according to the following formula: ;right and Calculate and operate according to the following formula: ;if, , then stop the tunnel fan; if , then let the tunnel fan run at full speed.

[0077] Optionally, the intelligent control of the tunnel fan according to the comparison result further includes: if , then calculate the power of the tunnel fan to be operated according to the following formula: ; In the formula, It is the power of the tunnel fan running at full speed.

[0078] according to The value of is used to control the speed of the tunnel fan using the PID control algorithm.

[0079] So far, the execution system has been explained. In actual operation, steps 2-1 to 2-18 can be repeated every 10 seconds according to the actual situation. It runs once every minute to achieve functions such as intelligent start, intelligent speed regulation, and intelligent stop.

[0080] Optionally, the intelligent control of the tunnel fan according to the comparison result further includes: performing equivalent calculation on the number of people and vehicles in the tunnel to obtain the equivalent number of people and vehicles, determining a reinforcement coefficient based on the equivalent number of people and vehicles, and using the reinforcement coefficient to adjust the power Make corrections and control the tunnel fan according to the corrected power run.

[0081] In this embodiment, the above embodiment calculates that In this case, the operating power of the tunnel fan is adjusted according to the presence of people and vehicles in the tunnel. However, this operating power does not take into account the presence of people and vehicles in the tunnel, and smoke will cause harm to people and vehicles in the tunnel.

[0082] Specifically, the number of people and vehicles in the tunnel is equivalently calculated to obtain the equivalent number of people and vehicles. Equivalent calculation refers to the equivalent conversion of people and vehicles. For example, the number of vehicles is converted into the number of people, and 1 car can be converted into 2 people. Then, the reinforcement coefficient is determined based on the equivalent number of people and vehicles. The reinforcement coefficient and the equivalent number of people and vehicles can be positively correlated. That is, the larger the equivalent number of people and vehicles in the tunnel, the larger the reinforcement coefficient is set. The reinforcement coefficient is a value greater than 1. Use the reinforcement coefficient to adjust the power Make corrections and control the tunnel fan according to the corrected power This setting can exhaust the smoke in the tunnel as quickly as possible to reduce the harm to people and vehicles in the tunnel.

[0083] like Figure 2 As shown, an embodiment of the present invention also discloses an intelligent control system for a tunnel fan based on smoke prediction, the system comprising a processing device and a storage device, the computer code stored in the storage device is called and executed by the processing device to implement the following steps: collecting and using training data to train the BiLSTM model; the sensor data acquisition module collects environmental data in the tunnel to obtain first smoke detector data, first temperature sensor data and first humidity sensor data, and simultaneously inputs the first smoke detector data, the first temperature sensor data and the first humidity sensor data into the trained BiLSTM model, the BiLSTM model outputs the predicted second smoke detector data, the second temperature sensor data and the second humidity sensor data at a future moment; the second smoke detector data is compared with the first interval, the second temperature sensor data is compared with the second interval, the second humidity sensor data is compared with the third interval, and the tunnel fan is intelligently controlled according to the comparison results; wherein the intelligent control includes intelligent start, intelligent speed regulation and intelligent stop.

[0084] An embodiment of the present invention further discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method described in the above embodiment.

[0085] An embodiment of the present invention further discloses a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in the above embodiment.

[0086] An embodiment of the present invention further discloses a computer program product, which includes computer code. When the computer code is executed by a processor of an electronic device, the method described in the above embodiment is implemented.

[0087] The computer-readable storage medium described above may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0088] 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.

[0089] 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 tunnel fan intelligent control method based on smoke prediction, characterized in that: The method comprises the following steps: collecting and using training data to train the BiLSTM model; using a sensor data acquisition module to acquire environmental data in the tunnel, and obtaining first smoke detector data, first temperature sensor data, and first humidity sensor data; inputting the first smoke detector data, the first temperature sensor data, and the first humidity sensor data into the trained BiLSTM model at the same time, and the BiLSTM model outputs the predicted second smoke detector data, second temperature sensor data, and second humidity sensor data at a future moment; comparing the second smoke detector data with the first interval, comparing the second temperature sensor data with the second interval, and comparing the second humidity sensor data with the third interval, and intelligently controlling the tunnel fan according to the comparison results; wherein the intelligent control comprises intelligent start, intelligent speed regulation, and intelligent stop; The intelligent control of the tunnel fan according to the comparison result includes: if the second smoke detector data reaches , then the tunnel fan is controlled to start running; if the second smoke detector data reaches , then the tunnel fan is controlled to run at full speed, that is, the smoke detector data is in the range When the second temperature sensor data reaches , then the tunnel fan is controlled to start running; if the second temperature sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the temperature sensor data is in the range When the tunnel fan is controlled to be in the running stage; if the second humidity sensor data reaches , then the tunnel fan is controlled to start running; if the second humidity sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the humidity sensor data is in the range When the tunnel fan is controlled to be in the operation stage; The intelligent control of the tunnel fan according to the comparison result also includes: setting the weight of the smoke detector value to , set the weight of the temperature sensor value to , set the weight of the humidity sensor value to ; Based on the first interval , the second interval and the third interval Calculate the comprehensive interval , calculated as follows: ; Calculate the average of each value in the second smoke detector data, the second temperature sensor data, and the second humidity sensor data respectively, that is, obtain ; right and Calculate and operate according to the following formula: ； if , then stop the tunnel fan; if , then let the tunnel fan run at full speed.

2. According to claim 1, a tunnel fan intelligent control method based on smoke prediction is characterized in that: The collecting and using training data to train the BiLSTM model includes: a sensor data acquisition module acquires environmental data in the tunnel, wherein the environmental data includes a third smoke detector data, a third temperature sensor data, and a third humidity sensor data; and based on the environmental data, a model initialization step, a training set generation step, and a training process step are respectively performed to realize the training of the BiLSTM model.

3. The intelligent control method for tunnel fans based on smoke prediction according to claim 1 is characterized in that: The intelligent control of the tunnel fan according to the comparison result also includes: if , then calculate the power of the tunnel fan to be operated according to the following formula: ; In the formula, is the power of the tunnel fan running at full speed; The value of is used to control the speed of the tunnel fan using the PID control algorithm.

4. The intelligent control method for tunnel fans based on smoke prediction according to claim 3 is characterized in that: The intelligent control of the tunnel fan according to the comparison result also includes: performing equivalent calculation on the number of people and vehicles in the tunnel to obtain the equivalent number of people and vehicles, determining a reinforcement coefficient based on the equivalent number of people and vehicles, and using the reinforcement coefficient to adjust the power Make corrections and control the tunnel fan according to the corrected power run.

5. An intelligent control system for tunnel fans based on smoke prediction, the system comprising a processing device and a storage device, characterized in that: The computer code stored in the storage device is called and executed by the processing device to implement the following steps: collect and use training data to train the BiLSTM model; the sensor data acquisition module collects environmental data in the tunnel to obtain first smoke detector data, first temperature sensor data and first humidity sensor data, and simultaneously inputs the first smoke detector data, the first temperature sensor data and the first humidity sensor data into the trained BiLSTM model, and the BiLSTM model outputs the predicted second smoke detector data, second temperature sensor data and second humidity sensor data at a future moment; compares the second smoke detector data with the first interval, compares the second temperature sensor data with the second interval, and compares the second humidity sensor data with the third interval, and intelligently controls the tunnel fan according to the comparison results; wherein the intelligent control includes intelligent start, intelligent speed regulation and intelligent stop; The intelligent control of the tunnel fan according to the comparison result includes: if the second smoke detector data reaches , then the tunnel fan is controlled to start running; if the second smoke detector data reaches , then the tunnel fan is controlled to run at full speed, that is, the smoke detector data is in the range When the second temperature sensor data reaches , then the tunnel fan is controlled to start running; if the second temperature sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the temperature sensor data is in the range When the tunnel fan is controlled to be in the running stage; if the second humidity sensor data reaches , then the tunnel fan is controlled to start running; if the second humidity sensor data reaches , then the tunnel fan is controlled to run at full speed, that is, the humidity sensor data is in the range When the tunnel fan is controlled to be in the operation stage; The intelligent control of the tunnel fan according to the comparison result also includes: setting the weight of the smoke detector value to , set the weight of the temperature sensor value to , set the weight of the humidity sensor value to ; Based on the first interval , the second interval and the third interval Calculate the comprehensive interval , calculated as follows: ; Calculate the average of each value in the second smoke detector data, the second temperature sensor data, and the second humidity sensor data respectively, that is, obtain ; right and Calculate and operate according to the following formula: ； if , then stop the tunnel fan; if , then let the tunnel fan run at full speed.

6. An electronic device comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 4.

7. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 4.

8. A computer program product, characterized in that: The computer program product includes computer codes, and when the computer codes are executed by a processor of an electronic device, the method according to any one of claims 1 to 4 is implemented.

Images