Method and system for controlling discharging temperature of asphalt mixing plant
By applying the LSTM deep learning model in the asphalt mixing station for temperature prediction and automatic heating temperature adjustment, the problem of hysteresis of discharge temperature control in the prior art is solved, and precise control of discharge temperature and improvement of production automation is achieved.
Patent Information
- Application Number
- CN202510031611.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-09
- Publication Date
- 2025-06-06
AI Technical Summary
The existing temperature control method for adjusting the heating power of the discharge temperature in asphalt mixing stations has serious hysteresis, resulting in unstable fluctuations in the discharge temperature.
The LSTM deep learning model is used to predict the temperature data in the asphalt mixing station, and the heating temperature of the drying cylinder is automatically adjusted to achieve accurate control of the discharge temperature.
By accurately predicting the discharge temperature, the demand and error of manual intervention is reduced, the automation level of the production process is improved, and the production efficiency and product quality of asphalt mixture are improved.
Smart Images
Figure CN120099834A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of discharge temperature control of an asphalt mixing station, and in particular to a discharge temperature control method and system of an asphalt mixing station. Background Art
[0002] At present, the demand for asphalt mixture in road construction is growing day by day. Asphalt mixing plant is a key link in road construction, and its production quality directly affects the quality and economic benefits of the whole project. Among them, the discharge temperature is a key indicator affecting the quality of asphalt mixture. At present, the discharge temperature is basically controlled manually. A trial mix is carried out before production. The operator adjusts the stone heating temperature (heating power) of the drying drum according to the trial mix temperature to make the discharge temperature meet the requirements. Because aggregates have different indicators such as density and moisture content, the temperature of the stone after drying with the same heating power will fluctuate, resulting in changes in the discharge temperature. The operator needs to adjust the heating power in real time according to the change of the discharge temperature. At the same time, because it takes 15 minutes to transfer the dried stone to the mixing bin, and because the material ratios of different gradations are different, the temporary storage time of stones of different specifications in the intermediate bin is different, which makes this temperature control method of adjusting the heating power according to the discharge temperature have serious lag. Summary of the invention
[0003] In view of the technical problems existing in the prior art, the purpose of the present invention is to provide a method for controlling the discharge temperature of an asphalt mixing plant, so as to solve the problem that the existing method for controlling the discharge temperature of an asphalt mixing plant by adjusting the heating power has serious hysteresis.
[0004] Another object of the present invention is to provide an asphalt mixing plant discharge temperature control system, which can automatically adjust the heating temperature according to the prediction results, thereby improving the automation level of the production process.
[0005] In order to achieve the above object, the present invention adopts the following technical solution: A method for controlling the discharge temperature of an asphalt mixing station comprises the following steps: S1: Obtain the temperature data of the drying drum, mixing bin, discharge port and asphalt bin in the asphalt mixing plant; S2: Convert temperature data into a format suitable for LSTM model input; S3: Set the target discharge temperature according to different asphalt mixtures; S4: Use the LSTM model to train the converted data to obtain the prediction result of the discharge temperature; S5: Determine whether the predicted result of the discharge temperature is equal to the target discharge temperature. If the predicted result of the discharge temperature is within the range of ±3°C of the target discharge temperature, the temperature of the drying drum will not be adjusted; if the predicted result of the discharge temperature is within the range of ±3-5°C of the target discharge temperature, the heating power of the drying drum will be adjusted; if the predicted result of the discharge temperature is outside the range of ±5°C of the target discharge temperature, a warning signal will be issued to prompt the operator to perform manual inspection.
[0006] As a preferred embodiment, in step S1, the method for obtaining temperature data of the drying drum, mixing bin, discharge port and asphalt bin is: installing temperature sensors on the drying drum and asphalt bin, installing infrared thermometers on the mixing bin and discharge port, and obtaining temperature data through the temperature sensors and infrared thermometers.
[0007] As a preferred embodiment, in step S1, the temperature data of the drying drum, mixing bin, discharge port and asphalt bin are collected in real time by the data acquisition and transmission module, and the temperature data is transmitted to the data processing module by wired or wireless means through the data acquisition and transmission module for preprocessing, and the temperature data is converted into a format suitable for LSTM model input.
[0008] As a preferred method, in step S2, the method of converting the temperature data into a format suitable for LSTM model input is: A: Use the data processing module to unify the unit of temperature data; B: Use the data processing module to arrange the temperature data in a time series format, with each time period corresponding to a set of characteristic values; C: De-noise and filter the temperature data for each time period, delete or fill in missing values, determine whether outliers are reasonable based on business logic, and remove unreasonable outliers; D: Normalize and scale the temperature data to eliminate the impact of the dimension on model training.
[0009] As a preferred embodiment, in step B, the temperature data is arranged in a time series format by dividing the data set according to time periods, using the data in each time period as a model input sample, and dividing the data set into a training set and a test set according to a ratio of 7:3.
[0010] As a preferred embodiment, in step S4, The LSTM model is used to train the converted data to obtain the prediction result of the discharge temperature: the data set converted from the temperature data of the drying drum, mixing bin, discharge port and asphalt bin is used as input, and the discharge temperature at the next moment is used as the target output; The LSTM model includes multiple LSTM layers, which are connected in sequence. Each LSTM layer receives the output of the previous time step and the input of the current time step. Each LSTM layer includes a forget gate, an input gate, and an output gate connected in sequence. The forget gate, the input gate, and the output gate are used to control the flow of information and extract long-term dependencies in the time series. The predicted discharge temperature value is output through the output gate.
[0011] As a preference, the LSTM layer uses Sigmoid as the activation function and selects mean square error as the loss function to measure the difference between the predicted temperature and the actual temperature.
[0012] As a preferred method, the LSTM model is trained using the training set data, and the model parameters are updated through the back propagation algorithm.
[0013] As a preference, set a suitable optimizer Adam, and adjust the learning rate and LSTM model parameters.
[0014] An asphalt mixing station discharge temperature control system, comprising: Temperature sensing module: The temperature sensing module is used to obtain the temperature of the drying drum, mixing bin, discharge port and asphalt bin; Data collection and transmission module: The data collection and transmission module is used to collect temperature data of the drying drum, mixing bin, discharge port and asphalt bin in real time, and send it to the temperature data processing module; A data processing module is used to convert temperature data into a format suitable for input into the LSTM model training module; Model training module, the model training module is used to train the converted data to obtain the prediction result of the discharge temperature; The control execution module compares the predicted result of the discharge temperature with the set target temperature, and determines whether the temperature of the drying cylinder needs to be adjusted according to the comparison result.
[0015] In general, the present invention has the following advantages: The method of the present invention uses the LSTM deep learning model to accurately predict the discharge temperature of the asphalt mixing plant, reducing the need for manual intervention and errors. The system can automatically adjust the heating temperature according to the prediction results, improving the automation level of the production process. By accurately controlling the discharge temperature, the production efficiency and product quality of the asphalt mixture are improved, and the scrap rate and cost are reduced. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 The figure is a schematic diagram of a method for controlling the discharge temperature of an asphalt mixing plant.
[0017] Figure 2This is a logical relationship diagram of an asphalt mixing plant discharge temperature control system.
[0018] Figure 3 A time relationship diagram of the data.
[0019] Figure 4 This is a schematic diagram of the LSTM deep learning model structure. DETAILED DESCRIPTION
[0020] The present invention will be further described in detail below in conjunction with specific implementation methods.
[0021] Embodiment 1 like Figure 1-Figure 4 As shown, the present embodiment provides a method for controlling the discharge temperature of an asphalt mixing station, comprising the following steps: S1: Obtain the temperature data of the drying drum, mixing bin, discharge port and asphalt bin in the asphalt mixing plant; S2: Convert temperature data into a format suitable for LSTM model input; S3: Set the target discharge temperature according to different asphalt mixtures; S4: Use the LSTM model to train the converted data to obtain the prediction result of the discharge temperature; S5: Determine whether the predicted result of the discharge temperature is equal to the target discharge temperature. If the predicted result of the discharge temperature is within the range of ±3°C of the target discharge temperature, the temperature of the drying drum will not be adjusted; if the predicted result of the discharge temperature is within the range of ±3-5°C of the target discharge temperature, the heating power of the drying drum will be adjusted; if the predicted result of the discharge temperature is outside the range of ±5°C of the target discharge temperature, a warning signal will be issued to prompt the operator to conduct manual inspection to prevent further problems caused by equipment failure or material abnormalities.
[0022] In step S1, the method for obtaining temperature data of the drying drum, mixing bin, discharge port and asphalt bin is: respectively installing PT100 temperature sensors on the drying drum and asphalt bin, installing infrared thermometers on the mixing bin and discharge port, and obtaining temperature data through the PT100 temperature sensor and infrared thermometer.
[0023] In step S1, the temperature data of the drying drum, mixing bin, discharge port and asphalt bin are collected in real time by the data acquisition and transmission module, and the temperature data is transmitted to the data processing module by wired or wireless means for cleaning and preprocessing, and the temperature data is converted into a format suitable for LSTM model input.
[0024] In step S2, the method for converting temperature data into a format suitable for LSTM model input is: A: Use the data processing module to unify the unit of temperature data; this system uses Celsius as the standard unit. The data processing module determines the temperature unit. If it is Celsius, it remains unchanged. If it is Fahrenheit, the system converts the Fahrenheit value into Celsius by subtracting 17.222.
[0025] B: Use the data processing module to arrange the temperature data in the format of a time series. The data collected by the data acquisition and transmission module includes the four temperatures of the drying drum, mixing bin, discharge port and asphalt bin. Each temperature value is an independent structure during collection and transmission, and contains information about time and temperature. The data processing module will process the data uniformly into a unified structure (time, drying drum temperature, mixing bin temperature, asphalt bin temperature and discharge port temperature). Among them, the method of arranging the temperature data in the format of a time series is: divide the data set according to time periods (such as every 30 minutes, every 60 minutes, etc.), and the data in each time period is used as a model input sample, and the data set is divided into a training set and a test set according to a ratio of 7:3.
[0026] C: Combined with business logic (such as the production process of the asphalt mixing plant, equipment restrictions, etc.), determine whether the abnormal value is reasonable and remove the unreasonable abnormal value; when the asphalt mixing plant is operating normally, the temperature of the drying drum, mixing bin, discharge port and asphalt bin should be between 100 and 300°C. Therefore, the temperature data below room temperature and above 500°C are judged as abnormal values and deleted. If one or more temperature values of the four temperature data are missing, the data will also be deleted.
[0027] D: Normalize and scale the temperature data to eliminate the impact of the dimension on model training. It is difficult to see the regularity of the temperature value changes in the original data. For example, the temperature of the drying drum is generally controlled at 200°C (different materials have different values). Assuming that the temperature is 195°C at a certain moment and 193°C at the next moment, the fluctuation pattern is not so clear from 195 to 193. By taking the first data of the time series as the benchmark, the temperature at subsequent moments is converted into a phase variable relative to the first data. For example, taking 195 as the benchmark and setting it to "0", 193 becomes "-1.03%", the fluctuation pattern is clearer and the model is easier to learn.
[0028] In step S4, the LSTM model is used to train the converted data to obtain the prediction result of the discharge temperature by taking the temperature of the drying drum, mixing bin, discharge port and asphalt bin for a period of time (the data set of step B in S2) as input, and the discharge temperature at the next moment as the target output; The LSTM model includes multiple LSTM layers, which are connected in sequence. Each LSTM layer receives the output of the previous time step and the input of the current time step. Each LSTM layer includes a forget gate, an input gate, and an output gate connected in sequence. The forget gate, the input gate, and the output gate are used to control the flow of information and extract the long-term dependency in the time series. The predicted discharge temperature value is output through the output gate.
[0029] The LSTM layer uses Sigmoid as the activation function; The mean square error (MSE) is chosen as the loss function to measure the difference between the predicted temperature and the actual temperature.
[0030] The LSTM model is trained with the training set data, and the model parameters are updated through the back-propagation algorithm to minimize the loss function.
[0031] Set the appropriate optimizer Adam and adjust the learning rate and LSTM model parameters.
[0032] During the training process, the trained LSTM model is evaluated using the test set data to verify its prediction accuracy. Based on the evaluation results, the model parameters are further adjusted and optimized to improve the prediction performance and achieve accurate prediction of the discharge temperature.
[0033] Embodiment 2 This embodiment provides an asphalt mixing station discharge temperature control system, including: Temperature sensing module: The temperature sensing module is used to obtain the temperature of the drying drum, mixing bin, discharge port and asphalt bin. The drying drum and asphalt bin use high-temperature resistant and high-precision PT100 temperature sensors suitable for their respective working environments to ensure accurate capture of temperature data. The mixing bin and discharge port use non-contact infrared thermal imagers (FOTRIC (616c-l47-412)) to quickly capture and analyze the temperature distribution of the asphalt mixture through thermal imaging technology. All temperature sensing devices are installed in the asphalt mixing plant and connected to the data processing module. Start the system, perform preliminary calibration and debugging, and ensure that all components are working properly.
[0034] Data acquisition and transmission module: The data acquisition and transmission module (signal separator, signal converter) is used to collect the temperature data of the drying drum, mixing bin, discharge port and asphalt bin in real time, and send it to the temperature data processing module; the data acquisition and transmission module is responsible for collecting the temperature data of each sensor in real time, and transmitting it to the data processing module by wire or wireless means. The data transmission process uses encryption technology to ensure data security and integrity.
[0035] A data processing module is used to convert temperature data into a format suitable for input into the LSTM model training module; The model training module is used to train the converted data to obtain the prediction result of the discharge temperature; the model training module adopts the LSTM model.
[0036] Control execution module: The control execution module compares the predicted result of the discharge temperature with the set target temperature, and determines whether to adjust the temperature of the drying drum based on the comparison result. According to the prediction result of the LSTM model, this module is responsible for automatically adjusting the heating temperature of the drying drum. When the predicted value is in the self-control area (i.e., within the range of ±3°C of the target temperature), the system automatically adjusts the heating power to maintain the stability of the discharge temperature. If the predicted value reaches the warning area (i.e., outside the range of ±5°C of the target temperature), the system automatically issues a warning signal and prompts the operator to intervene and check to prevent further problems caused by equipment failure or material abnormalities.
[0037] The parts not mentioned in this embodiment are the same as those in the first embodiment.
[0038] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.
Claims
1. A method for controlling the discharge temperature of an asphalt mixing station, characterized in that: The following steps are involved: S1: Obtain the temperature data of the drying drum, mixing bin, discharge port and asphalt bin in the asphalt mixing plant; S2: Convert temperature data into a format suitable for LSTM model input; S3: Set the target discharge temperature according to different asphalt mixtures; S4: Use the LSTM model to train the converted data to obtain the prediction result of the discharge temperature; S5: Determine whether the predicted result of the discharge temperature is equal to the target discharge temperature. If the predicted result of the discharge temperature is within the range of ±3°C of the target discharge temperature, the temperature of the drying drum will not be adjusted; if the predicted result of the discharge temperature is within the range of ±3-5°C of the target discharge temperature, the heating power of the drying drum will be adjusted; if the predicted result of the discharge temperature is outside the range of ±5°C of the target discharge temperature, a warning signal will be issued to prompt the operator to perform manual inspection.
2. A method for controlling the discharge temperature of an asphalt mixing station according to claim 1, characterized in that: In step S1, the method for obtaining temperature data of the drying drum, mixing bin, discharge port and asphalt bin is: installing temperature sensors on the drying drum and asphalt bin, installing infrared thermometers on the mixing bin and discharge port, and obtaining temperature data through the temperature sensors and infrared thermometers.
3. A method for controlling the discharge temperature of an asphalt mixing station according to claim 2, characterized in that: In step S1, the temperature data of the drying drum, mixing bin, discharge port and asphalt bin are collected in real time by the data acquisition and transmission module, and the temperature data is transmitted to the data processing module by wired or wireless means for preprocessing through the data acquisition and transmission module, and the temperature data is converted into a format suitable for LSTM model input.
4. The method for controlling the discharge temperature of an asphalt mixing station according to claim 1, characterized in that: In step S2, the method for converting temperature data into a format suitable for LSTM model input is: A: Use the data processing module to unify the unit of temperature data; B: Use the data processing module to arrange the temperature data in a time series format, with each time period corresponding to a set of characteristic values; C: De-noise and filter the temperature data for each time period, delete or fill in missing values, determine whether outliers are reasonable based on business logic, and remove unreasonable outliers; D: Normalize and scale the temperature data to eliminate the impact of the dimension on model training.
5. A method for controlling the discharge temperature of an asphalt mixing station according to claim 4, characterized in that: In step B, the temperature data is arranged in a time series format by dividing the data set according to time periods, using the data in each time period as a model input sample, and dividing the data set into a training set and a test set according to a ratio of 7:
3.
6. A method for controlling the discharge temperature of an asphalt mixing station according to claim 5, characterized in that: In step S4, the LSTM model is used to train the converted data to obtain the prediction result of the discharge temperature by taking the data set converted from the temperature data of the drying drum, the mixing bin, the discharge port and the asphalt bin as input, and the discharge temperature at the next moment as the target output; The LSTM model includes multiple LSTM layers, which are connected in sequence. Each LSTM layer receives the output of the previous time step and the input of the current time step. Each LSTM layer includes a forget gate, an input gate, and an output gate connected in sequence. The forget gate, the input gate, and the output gate are used to control the flow of information and extract the long-term dependency in the time series. The predicted discharge temperature value is output through the output gate.
7. A method for controlling the discharge temperature of an asphalt mixing station according to claim 6, characterized in that: The LSTM layer uses Sigmoid as the activation function; The mean squared error is chosen as the loss function to measure the difference between the predicted temperature and the actual temperature.
8. A method for controlling the discharge temperature of an asphalt mixing station according to claim 6, characterized in that: The LSTM model is trained with the training set data, and the model parameters are updated through the back propagation algorithm.
9. A method for controlling the discharge temperature of an asphalt mixing station according to claim 6, characterized in that: Set the appropriate optimizer Adam and adjust the learning rate and LSTM model parameters.
10. An asphalt mixing station discharge temperature control system, characterized in that: include: Temperature sensing module: The temperature sensing module is used to obtain the temperature of the drying drum, mixing bin, discharge port and asphalt bin; Data collection and transmission module: The data collection and transmission module is used to collect temperature data of the drying drum, mixing bin, discharge port and asphalt bin in real time, and send it to the temperature data processing module; A data processing module is used to convert temperature data into a format suitable for input into the LSTM model training module; Model training module, the model training module is used to train the converted data to obtain the prediction result of the discharge temperature; The control execution module compares the predicted result of the discharge temperature with the set target temperature, and determines whether to adjust the temperature of the drying cylinder according to the comparison result.