Start-up control method and device of industrial heating equipment, cloud server and storage medium

Abstract

The invention discloses a startup control method and device of industrial heating equipment, a cloud server and a storage medium. The method comprises the following steps: acquiring equipment waste heat of the industrial heating equipment sent by an Internet of Things platform; the residual temperature of the equipment is input into a two-channel hybrid prediction model, the temperature rise prediction duration is obtained, and the model is obtained after a main channel regression model is compensated through a dynamic compensation auxiliary channel; and performing startup control on the equipment according to the temperature rise prediction duration. Through the above mode, the temperature rise duration can be accurately predicted, the problem of too long start-up in advance caused by artificial experience is avoided, the energy consumption is greatly reduced, meanwhile, the dual-channel hybrid prediction model obtained by compensating the main channel regression model through the dynamic compensation auxiliary channel is adopted, the accumulated error is reduced, and the prediction accuracy is improved. The system gets rid of the limitation of localized deployment by means of an Internet of Things platform, reduces the deployment cost, achieves the data sharing of multiple workshops, breaks through the data isolated island, and improves the efficiency and convenience of production management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of control, in particular to a start-up control method and device of an industrial heating device, a cloud server and a storage medium. BACKGROUND

[0002] In industrial production, high-energy-consumption devices such as drying rooms and RTO (Regenerative Thermal Oxidizer) furnaces need to be started up in advance for preheating. At present, the start-up time is mainly determined by manual experience, resulting in serious energy waste. The existing preheating control scheme mainly uses a neural network or a random forest model to predict the temperature rising time, but such a model requires a large amount of data and computing power, and has problems such as overfitting, large cumulative error and poor interpretability. It is difficult for technicians to understand the prediction logic, which is not conducive to production adjustment and optimization, and is difficult to adapt to sudden changes in working conditions, and cannot meet the long-term stable prediction demand. SUMMARY

[0003] The main purpose of the present application is to provide a start-up control method and device of an industrial heating device, a cloud server and a storage medium, aiming to solve the technical problems of serious energy waste, poor interpretability of the prediction model and large cumulative error in the preheating control scheme of high-energy-consumption devices in the prior art.

[0004] To achieve the above-mentioned purpose, the present application provides a start-up control method of an industrial heating device, which is applied to a cloud server, the cloud server is in communication connection with an Internet of Things platform, and the method comprises the following steps:

[0005] obtaining the device residual temperature of the industrial heating device sent by the Internet of Things platform;

[0006] inputting the device residual temperature into a double-channel hybrid prediction model, performing time prediction through the double-channel hybrid prediction model, and obtaining the temperature rising prediction time of the industrial heating device, wherein the double-channel hybrid prediction model is obtained by compensating a main channel regression model through a dynamic compensation auxiliary channel;

[0007] controlling the start-up of the industrial heating device according to the temperature rising prediction time.

[0008] In an embodiment, before the step of inputting the device residual temperature into a double-channel hybrid prediction model, performing time prediction through the double-channel hybrid prediction model, and obtaining the temperature rising prediction time of the industrial heating device, the method further comprises the following steps:

[0009] performing temperature rising rate analysis according to the historical temperature rising data of the industrial heating device, and determining the device temperature rising rate;

[0010] The device temperature rise rate, device purging time, target device temperature, preset temperature coefficient, and residual temperature variable are used to construct a model to obtain a main channel regression model;

[0011] The dynamic compensation auxiliary channel is used to compensate and analyze the temperature deviation data of the industrial heating device in a historical window to determine a dynamic compensation coefficient;

[0012] The main channel regression model is compensated by the dynamic compensation coefficient to obtain a double-channel hybrid prediction model.

[0013] In an embodiment, the step of determining the device temperature rise rate according to historical temperature rise data of the industrial heating device comprises:

[0014] The historical temperature rise data of the industrial heating device are used to determine a start-up temperature corresponding to each control period, a start-up time corresponding to each control period, a temperature rise critical time corresponding to each control period, and a temperature rise standard time corresponding to each control period in a historical time period;

[0015] The start-up temperature corresponding to each control period, the start-up time corresponding to each control period, the temperature rise standard time corresponding to each control period, and the target device temperature corresponding to each control period are used to calculate a temperature rise rate to obtain a basic temperature rise rate of the industrial heating device;

[0016] The temperature rise critical time corresponding to each control period, the temperature rise standard time corresponding to each control period, the target device temperature corresponding to each control period, and a target critical temperature corresponding to each control period are used to calculate a temperature rise rate to obtain a high-temperature temperature rise rate of the industrial heating device;

[0017] The basic temperature rise rate and the high-temperature temperature rise rate are used to obtain a device temperature rise rate.

[0018] In an embodiment, the step of calculating a temperature rise rate according to the start-up temperature corresponding to each control period, the start-up time corresponding to each control period, the temperature rise standard time corresponding to each control period, and the target device temperature corresponding to each control period to obtain a basic temperature rise rate of the industrial heating device comprises:

[0019] The start-up temperature corresponding to each control period and the target device temperature corresponding to each control period are used to calculate a temperature difference to obtain a temperature difference value corresponding to each control period;

[0020] The start-up time corresponding to each control period and the temperature rise standard time corresponding to each control period are used to calculate a time difference to obtain a time difference value corresponding to each control period;

[0021] The rate calculation is performed based on the time difference corresponding to each control period and the temperature difference corresponding to each control period to obtain the heating rate corresponding to each control period;

[0022] The heating rates corresponding to each control period are averaged to obtain the basic heating rate of the industrial heating equipment.

[0023] In one embodiment, the step of performing compensation deviation analysis on the temperature deviation data of the industrial heating equipment within the historical window according to the dynamic compensation auxiliary channel to determine the dynamic compensation coefficient includes:

[0024] performing variance calculation on the temperature deviation data of the industrial heating equipment within a historical window according to the dynamic compensation auxiliary channel to obtain the temperature deviation variance of the industrial heating equipment;

[0025] Modeling is performed based on the temperature deviation variance, the adaptive coefficient, and the off-diagonal element matrix to obtain a dynamic deviation index;

[0026] A dynamic compensation coefficient is determined according to the dynamic deviation index and the Kalman filter algorithm.

[0027] In one embodiment, the step of determining the dynamic compensation coefficient according to the dynamic deviation index and the Kalman filter algorithm includes:

[0028] A state equation is constructed according to the dynamic deviation index and the historical compensation coefficient to obtain the deviation compensation coefficient;

[0029] An observation equation is constructed according to the deviation compensation coefficient and the target observation noise to obtain a target observation value;

[0030] Perform prediction node calculation based on the historical error covariance and the dynamic deviation index to obtain the prediction error covariance;

[0031] Performing a Kalman gain calculation based on the prediction error covariance, the target observation noise, and the historical error covariance to obtain a Kalman gain;

[0032] A coefficient is calculated based on the target observation value, the Kalman gain and the predicted state estimate to obtain a dynamic compensation coefficient.

[0033] In one embodiment, the step of controlling the startup of the industrial heating equipment according to the predicted temperature rise time includes:

[0034] Determining a predicted startup time of the industrial heating equipment according to the target working time and the predicted temperature rise time;

[0035] Push the predicted boot time to the administrator's terminal.

[0036] In one embodiment, after the step of controlling the startup of the industrial heating equipment according to the predicted temperature rise time, the method further includes:

[0037] Obtaining the actual device temperature of the industrial heating equipment during the target working time;

[0038] calculating an actual temperature deviation between a target device temperature and the actual device temperature;

[0039] When the actual temperature deviation is not within the preset deviation range, the temperature warning information is pushed to the administrator's terminal for deviation warning.

[0040] In addition, to achieve the above-mentioned purpose, the present application also proposes a startup control device for industrial heating equipment, the startup control device for industrial heating equipment comprising:

[0041] The acquisition module is used to obtain the residual temperature of industrial heating equipment sent by the IoT platform;

[0042] a prediction module, configured to input the residual temperature of the equipment into a dual-channel hybrid prediction model, perform time prediction using the dual-channel hybrid prediction model, and obtain a predicted temperature rise duration of the industrial heating equipment, wherein the dual-channel hybrid prediction model is obtained by compensating the main channel regression model through a dynamic compensation auxiliary channel;

[0043] A control module is used to control the startup of the industrial heating equipment according to the predicted temperature rise time.

[0044] In addition, to achieve the above-mentioned purpose, the present application also proposes a cloud server, which includes: a memory, a processor, and a computer program stored on the memory and runnable on the processor, wherein the computer program is configured to implement the steps of the startup control method of the industrial heating equipment as described above.

[0045] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the power-on control method of the industrial heating equipment as described above are implemented.

[0046] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the startup control method of the industrial heating equipment as described above.

[0047] The method of the application is applied to a cloud server in communication connection with an Internet of Things platform, and the method comprises: obtaining device residual temperature of an industrial heating device sent by the Internet of Things platform; inputting the device residual temperature into a double-channel hybrid prediction model to perform time prediction through the double-channel hybrid prediction model to obtain a heating prediction duration of the industrial heating device, the double-channel hybrid prediction model being obtained by compensating a main channel regression model through a dynamic compensation auxiliary channel; and performing start-up control on the industrial heating device according to the heating prediction duration. In the foregoing manner, by obtaining the industrial heating device residual temperature of the Internet of Things platform, inputting the double-channel hybrid prediction model to obtain the heating prediction duration, and controlling the device start-up accordingly, the heating duration can be accurately predicted, the problem of excessively long start-up caused by manual experience is avoided, energy consumption is greatly reduced, the double-channel hybrid prediction model obtained by compensating the main channel regression model through the dynamic compensation auxiliary channel is used, the interpretability of the model is enhanced, the model can adapt to sudden changes in working conditions, and cumulative errors are reduced; and the localization deployment limit is eliminated with the aid of the Internet of Things platform, the deployment cost is reduced, multi-workshop data sharing is realized, data islands are broken, and production management efficiency and convenience are improved. BRIEF DESCRIPTION OF DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application together with the specification.

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced in the following. Obviously, for those skilled in the field, other drawings can also be obtained based on these drawings without creative labor.

[0050] Figure 1 A flowchart is provided for the start-up control method of the industrial heating device in embodiment one of the application;

[0051] Figure 2 A system architecture diagram is provided for the start-up control method of the industrial heating device in embodiment one of the application;

[0052] Figure 3 A prediction workflow diagram is provided for the start-up control method of the industrial heating device in embodiment one of the application;

[0053] Figure 4 A flowchart is provided for the start-up control method of the industrial heating device in embodiment two of the application;

[0054] Figure 5 A flowchart is provided for the start-up control method of the industrial heating device in embodiment three of the application;

[0055] Figure 6 A module structure diagram of the start-up control device of the industrial heating equipment in the embodiment of the present application is shown in Figure 1.

[0056] Figure 7 A cloud server structure diagram of the hardware running environment involved in the start-up control method of the industrial heating equipment in the embodiment of the present application is shown in Figure 2.

[0057] The object realization, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0058] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0059] In order to better understand the technical solutions of the present application, the specific embodiments will be described in detail below with reference to the drawings and the accompanying drawings.

[0060] The main solution of the embodiment of the present application is: obtaining the device residual temperature of the industrial heating equipment sent by the Internet of Things platform; inputting the device residual temperature into a double-channel hybrid prediction model, performing time prediction through the double-channel hybrid prediction model to obtain the temperature rise prediction time length of the industrial heating equipment, the double-channel hybrid prediction model being obtained by compensating a main channel regression model through a dynamic compensation auxiliary channel; and performing start-up control on the industrial heating equipment according to the temperature rise prediction time length.

[0061] In industrial production, high-energy-consumption equipment such as drying rooms and RTO furnaces need to be started in advance for preheating. At present, the start-up time is mainly determined by manual experience, resulting in waste of energy consumption. Taking a resin workshop of a company as an example, the temperature rise of the 24-year-old drying room equipment to the position time is 12 minutes earlier than the formal production time on average, and the RTO furnace is 35 minutes earlier, with an annual power loss of 180,000 kWh and a gas loss of 28,000 m 3 The existing preheating control scheme mainly uses neural network or random forest model for temperature rise time prediction, but such models require a large amount of data and computing power, and have problems of overfitting, large cumulative error and poor interpretability, which makes it difficult for technicians to understand the prediction logic, is not conducive to production adjustment and optimization, and is difficult to adapt to sudden changes in working conditions, and cannot meet the long-term stable prediction demand. At the same time, the existing scheme is deployed locally, using a "PLC + local server + desktop display screen" architecture, with a single workshop deployment cost ≥ 100,000 yuan, data islands exist, and deployment costs are increased. In addition, local deployment cannot share data across workshops, is inconvenient to operate and cannot realize mobile terminal notification, limiting production management efficiency and convenience.

[0062] The application provides a solution, by obtaining the industrial heating equipment residual temperature of the Internet of Things platform, inputting a double-channel mixed prediction model to obtain the temperature rising prediction duration, and controlling the equipment start-up according to the temperature rising prediction duration, the temperature rising duration can be accurately predicted, the problem of overlong start-up caused by artificial experience is avoided, the energy consumption is greatly reduced, and the double-channel mixed prediction model obtained by compensating the main channel regression model by the dynamic compensation auxiliary channel is used, so that the interpretability of the model is enhanced, the cumulative error is reduced, the localization deployment limit is broken through the Internet of Things platform, the deployment cost is reduced, the multi-workshop data sharing is realized, the data island is broken, and the production management efficiency and convenience are improved.

[0063] It should be noted that the execution subject of the embodiment can be a computing service cloud server with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a cloud server capable of realizing the above functions. The following takes the cloud server as an example to illustrate the embodiment and the following embodiments.

[0064] Based on this, the application embodiment provides an industrial heating equipment start-up control method, referring to Figure 1 , Figure 1 The flowchart of the first embodiment of the industrial heating equipment start-up control method of the application is shown.

[0065] In the embodiment, the method is applied to a cloud server, the cloud server is in communication connection with an Internet of Things platform, and the industrial heating equipment start-up control method includes steps S10-S40.

[0066] Step S10, obtaining the equipment residual temperature of the industrial heating equipment sent by the Internet of Things platform.

[0067] It should be noted that the method of the embodiment adopts a three-layer full-cloud service chain architecture of cross-platform collaboration, uses the Internet of Things platform to collect the equipment temperature, environmental temperature and humidity data of the industrial heating equipment in real time, stores them into a time sequence database, and supports API (Application Programming Interface, application programming interface) calling. At the same time, the related historical information of the equipment is stored in the multi-dimensional table of the enterprise collaboration and management platform, and is integrated and interacted in the cloud server through the API.

[0068] It is understandable that through a fully cloud-based service chain collaborative architecture, the entire process from data collection, model calculation, and mobile terminal push is cloud-based, and a lightweight architecture design with cross-platform systematic integration has achieved full cloud-based service chain integration from the data layer to the model layer to the control layer, breaking through the limitations of traditional local deployment and realizing multi-workshop data collaborative processing and automatic mobile push of startup information. This solves the core defects of traditional solutions such as high local deployment costs and data silos. Among them, the data layer includes the IoT platform's time series database and the dynamic database of the enterprise collaboration and management platform, supporting API data exchange; the model layer includes the predictive model workflow and deviation detection workflow deployed on the cloud server, automatically implementing multiple core functions including data call, dynamic compensation calculation, regression model calculation, message generation, deviation recording, etc. The application layer triggers the workflow on a regular basis through the low-code development platform and pushes the startup command to the user's terminal via the messaging platform.

[0069] In the specific implementation, take industrial heating equipment as drying room, RTO and other high energy consumption equipment, and cloud server as enterprise large model platform as an example. Figure 2 As shown, Figure 2 The architecture demonstrates the collaborative relationship and full-process coordination between various components: data collection on the IoT platform, a time-series database, model workflow deployment on the large-scale model platform, scheduled workflow triggering on the low-code development platform, and real-time push notifications via instant messaging or an interactive terminal tool (i.e., App 1). Specifically, the PLC (Programmable Logic Controller) performs local control and data collection for industrial heating equipment, acquiring real-time operating parameters. The collection server, serving as industrial data collection middleware, connects to the PLC, standardizes and aggregates raw equipment data, and transmits it to the IoT platform via industrial communication protocols, ensuring real-time and accurate data. The IoT platform receives device data from the collection server and, leveraging the characteristics of the time-series database, stores high-frequency, time-series data such as temperature and humidity in a time series format. This allows for rapid query and review of historical equipment operating status, and provides data access interfaces for the upper-level model and application layers, enabling data access via APIs. The large-scale model platform deploys a dual-channel hybrid prediction model. Leveraging data from the data layer and historical information provided by the enterprise collaboration and management platform (i.e., App 2), it predicts the heating time of industrial heating equipment and pushes the prediction results to the front-end. The front desk faces the end users and presents information such as the equipment operation status and the prediction results of the large model platform through instant messaging / interactive terminals such as APP1.

[0070] It should be noted that the industrial heating equipment is an industrial production equipment that needs to be precisely controlled in temperature, such as an oven and an RTO furnace. The equipment residual temperature T1 refers to the temperature value retained by the industrial heating equipment after the industrial heating equipment stops heating. The equipment residual temperature is collected in real time by an Internet of Things platform and uploaded to a cloud server. When the cloud server triggers the temperature rise duration prediction of the industrial heating equipment, the equipment residual temperature at the triggering time of the prediction model workflow sent by the Internet of Things platform is obtained.

[0071] In step S20, the equipment residual temperature is input into the dual-channel hybrid prediction model, and the temperature rise prediction duration of the industrial heating equipment is obtained by time prediction through the dual-channel hybrid prediction model. The dual-channel hybrid prediction model is obtained by compensating the main channel regression model through a dynamic compensation auxiliary channel.

[0072] It should be noted that the dual-channel hybrid prediction model is a composite prediction model that fuses the main channel regression model and the dynamic compensation auxiliary channel. Through dual-path cooperation, the accuracy and adaptability of the temperature rise duration prediction of the industrial heating equipment are improved. The main channel regression model is a basic prediction model of input parameters-temperature rise duration, which is constructed based on the historical data of the industrial heating equipment through a regression algorithm (such as linear regression, decision tree regression). The dynamic compensation auxiliary channel introduces a real-time dynamic compensation coefficient to correct the main channel prediction for the error blind area (such as residual temperature fluctuation, environmental interference, and complex scenes) of the main channel model, forming a dual-path cooperative mechanism of basic prediction and dynamic compensation.

[0073] It can be understood that inputting the equipment residual temperature into the trained dual-channel hybrid prediction model can output the required duration of the industrial heating equipment from the equipment residual temperature to the target equipment temperature through the dual-channel hybrid prediction model. The target equipment temperature refers to the ideal temperature value that the industrial heating equipment should reach to meet specific production requirements. The temperature rise prediction duration refers to the required duration of the industrial heating equipment from the equipment residual temperature to the target equipment temperature.

[0074] In step S30, the industrial heating equipment is controlled according to the temperature rise prediction duration.

[0075] It should be noted that the cloud server generates a start-up time suggestion in combination with the temperature rise prediction duration and the scheduled time point at which the industrial heating equipment needs to formally start performing a production task, and pushes a start-up notification message to a terminal of a device start-up control personnel, thereby realizing precise prediction and timely notification of the start-up time of the equipment. Alternatively, the cloud server issues a start-up time to the industrial heating equipment, and the industrial heating equipment automatically starts up according to the start-up time. However, in this embodiment, the start-up of the industrial heating equipment is realized by manual operation for safety consideration.

[0076] In a possible implementation, step S30 can further include steps A11-A12.

[0077] Step A11, determining the predicted start-up time of the industrial heating equipment according to the target working time and the temperature rise prediction duration.

[0078] Step A12, pushing the predicted start-up time to the terminal of the administrator.

[0079] It should be noted that the target working time refers to the scheduled time point at which the industrial heating equipment needs to formally start performing a production task. In order to make the industrial heating equipment reach the target equipment temperature on time at the target working time and start working, the time point at which the industrial heating equipment needs to start heating is calculated through the temperature rise prediction duration. In this embodiment, the predicted start-up time refers to the time point at which the industrial heating equipment needs to start heating, and the predicted start-up time = target working time - temperature rise prediction duration.

[0080] It can be understood that the cloud server pushes the related notification message of the predicted start-up time to the terminal of the administrator, realizes accurate prediction and timely notification of the equipment start-up time, and the administrator performs start-up operation on the industrial heating equipment according to the predicted start-up time.

[0081] In a possible implementation, after step S30, steps B11-B13 can be further included.

[0082] Step B11, obtaining the actual equipment temperature of the industrial heating equipment at the target working time.

[0083] Step B12, calculating the actual temperature deviation between the target equipment temperature and the actual equipment temperature.

[0084] Step B13, when the actual temperature deviation is not within the preset deviation range, pushing temperature warning information to the terminal of the administrator for deviation warning.

[0085] It should be noted that after the start-up heating operation is performed on the industrial heating equipment, the deviation detection workflow in the cloud server is triggered at a fixed time, for example, the deviation detection workflow is triggered at 8:30 every morning, the actual equipment temperature of the industrial heating equipment at the target working time is obtained, and the actual temperature deviation between the target equipment temperature and the actual equipment temperature is calculated. In this embodiment, the actual temperature deviation F1 = target equipment temperature - actual equipment temperature.

[0086] It is understandable that after obtaining the actual temperature deviation, since the actual temperature deviation is the core indicator for measuring whether the equipment has reached a production-ready state, excessive deviation will directly affect product quality. Therefore, the actual temperature deviation is compared with the preset deviation range to determine whether the actual temperature deviation is within the preset deviation range. If the actual temperature deviation is within the preset deviation range, it means that the actual temperature deviation is within the limit at this time. In this case, the actual temperature deviation is only saved to the dynamic database in the enterprise collaboration and management platform for subsequent model calculations to achieve adaptive dynamic optimization of the model, and the actual temperature deviation is pushed to the administrator's terminal for display, without a deviation warning.

[0087] In the specific implementation, if the actual temperature deviation is not within the preset deviation range, it means that the actual temperature deviation exceeds the limit, which triggers an early warning and pushes the temperature warning information to the administrator's terminal, prompting the administrator to confirm and correct the model parameters. The temperature warning information includes but is not limited to information that the actual temperature deviation is not within the preset deviation range and the specific data value of the actual temperature deviation. In addition to pushing the temperature warning information to the administrator's terminal, the actual temperature deviation must also be saved to the dynamic database in the enterprise collaboration and management platform for subsequent model calculations to achieve adaptive dynamic optimization of the model.

[0088] It should be noted that the prediction model workflow and deviation detection workflow are triggered regularly through the low-code development platform. For example, the prediction model workflow is triggered regularly at 6:30 every day to execute the following steps: obtaining the residual temperature of the device - model calculation - generating a boot time recommendation, and pushing a message to the administrator's terminal through instant messaging or interactive terminal tools; the deviation detection workflow is triggered regularly at 8:30 every day to execute the following steps: obtaining the actual device temperature - deviation calculation - saving the deviation in the dynamic database. The specific process of the prediction model workflow is as follows: Figure 3 As shown, the residual temperature of the equipment is obtained through the Internet of Things platform, and the temperature T2 is calculated using the residual temperature of the equipment. The basic deletion is calculated using linear regression, and the auxiliary channel process is compensated: historical error - Kalman filter - generate the current deviation compensation F, and the current deviation compensation is injected into the main channel to participate in the calculation, and the temperature rise prediction duration is obtained. A power-on time recommendation is generated, and the power-on time is pushed through instant messaging or interactive terminal tools to realize dynamic deviation compensation logic and dual-channel data fusion.

[0089] The method of the embodiment is applied to a cloud server in communication connection with an Internet of Things platform, and the method comprises: obtaining device residual temperature of an industrial heating device sent by the Internet of Things platform; inputting the device residual temperature into a double-channel hybrid prediction model to perform time prediction through the double-channel hybrid prediction model to obtain a heating prediction duration of the industrial heating device, the double-channel hybrid prediction model being obtained by compensating a main channel regression model through a dynamic compensation auxiliary channel; and performing start-up control on the industrial heating device according to the heating prediction duration. In the foregoing manner, the device residual temperature of the Internet of Things platform is obtained, the double-channel hybrid prediction model is inputted to obtain the heating prediction duration, and the device is controlled to start up according to the heating prediction duration, so that the heating duration can be accurately predicted, the problem of overlong start-up due to manual experience is avoided, energy consumption is greatly reduced, the double-channel hybrid prediction model obtained by compensating the main channel regression model through the dynamic compensation auxiliary channel is adopted, the interpretability of the model is enhanced, the model can adapt to sudden changes in working conditions, and cumulative error is reduced. The Internet of Things platform is relied on to get rid of the limitation of local deployment, deployment cost is reduced, multi-workshop data sharing is realized, a data island is broken, and production management efficiency and convenience are improved.

[0090] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as the above embodiment one can refer to the above introduction, and the following will not be described in detail. On this basis, please refer to Figure 4 , in the start-up control method of the industrial heating device, before the step S20, further comprising steps S21-S24:

[0091] Step S21, performing heating rate analysis according to historical heating data of the industrial heating device to determine the device heating rate.

[0092] It should be noted that the historical heating data includes but is not limited to the temperature change data of the industrial heating device with time in each control period in the past period. The device heating rate includes a basic heating rate K1 and a high-temperature zone heating rate K2. In the embodiment, the control period is set to one day, which can also be adjusted according to requirements. The unit of the device heating rate is ℃ / min, which can also be adjusted according to requirements. The embodiment does not limit this. In the embodiment, the past period is a historical interval with the current time as the end of the time interval and a fixed interval length (for example, 10 days) as the time step, which can also be flexibly configured according to the device running period and process requirements. The embodiment does not limit this.

[0093] It can be understood that the basic heating rate refers to the average heating speed of the industrial heating device from the device residual temperature to the target device temperature; and the high-temperature zone heating rate refers to the average heating speed of the industrial heating device from the set high-temperature critical value to the target device temperature.

[0094] In a specific implementation, for the temperature change data of each control period in the historical temperature rise data, the temperature rise speed of the industrial heating equipment from the equipment residual temperature to the target equipment temperature and the temperature rise speed of the industrial heating equipment from the set high temperature critical value to the target equipment temperature are calculated respectively, the average of the temperature rise speed of the industrial heating equipment from the equipment residual temperature to the target equipment temperature in all control periods is calculated to obtain the basic temperature rise rate K1, and the average of the temperature rise speed of the industrial heating equipment from the set high temperature critical value to the target equipment temperature in all control periods is calculated to obtain the basic temperature rise rate K2.

[0095] In step S22, a model is constructed according to the equipment temperature rise rate, the equipment purging time, the target equipment temperature, the preset temperature coefficient, and the residual temperature variable to obtain a main channel regression model.

[0096] It should be noted that the equipment purging time S refers to a fixed time length S of purging operation performed before heating to ensure production safety and remove residual medium in the equipment; the preset temperature coefficient is a fixed parameter used to calculate the equipment temperature after the purging operation is completed, and in this embodiment, the preset temperature coefficient includes two fixed parameters a and b, where a is set to 0.7312 and b is set to 5.8107, which can also be adjusted according to requirements, and this embodiment does not limit this.

[0097] It can be understood that the residual temperature variable T1 is a key variable affecting the temperature rise time, and when predicting the temperature rise time before each start of the industrial heating equipment, the specific value of the residual temperature variable is the equipment residual temperature T1 sent by the Internet of Things platform. The main channel regression model is obtained by constructing a model using the equipment temperature rise rate, the equipment purging time, the target equipment temperature, the preset temperature coefficient, and the residual temperature variable, and in this embodiment, the main channel regression model adopts a multi-segment linear regression model, and the specific formula is: T=(T0-T2)×K1+S+F×K2, where T2=a×T1+b is the equipment temperature after the purging operation is completed; T0 is the target equipment temperature, for example, the target equipment temperature of the drying room can be set to 88℃; and F is a dynamic compensation coefficient.

[0098] In a specific implementation, since K1 and K2 are affected by historical temperature data, and the dynamic compensation coefficient is also dynamically changed, the main channel regression model corresponding to each control period is also dynamically updated and adaptively adjusted to match the change of the temperature rise characteristics of the industrial heating equipment caused by long-term operation, working condition fluctuation, and other factors, so as to ensure that the model always fits the actual heating law of the equipment.

[0099] In step S23, the temperature deviation data of the industrial heating equipment in the historical window is compensated and analyzed according to the dynamic compensation auxiliary channel to determine the dynamic compensation coefficient.

[0100] It should be noted that the dynamic compensation auxiliary channel is independent of the error correction module of the main channel, and in the embodiment, the dynamic compensation auxiliary channel uses a dynamic Q value Kalman filtering algorithm based on residual driving to optimize the prediction deviation of the main channel regression model in real time.

[0101] It can be understood that the history window refers to a pre-set fixed time window, for example, the last 7 days. The temperature deviation data includes the temperature difference corresponding to each control period within the history window, which is obtained by subtracting the actual device temperature at the target working time from the target device temperature.

[0102] In a specific implementation, after the temperature deviation data within the history window is filtered and corrected by the dynamic compensation auxiliary channel, a dynamic compensation coefficient K corresponding to the current control period is output, which is used to adjust the prediction result of the main channel regression model and improve its adaptive ability.

[0103] In step S24, the main channel regression model is compensated by the dynamic compensation coefficient to obtain a dual-channel hybrid prediction model.

[0104] It should be noted that the dynamic compensation coefficient K corresponding to the current control period calculated by the dynamic compensation auxiliary channel is substituted into the main channel regression model to correct the main channel regression model, thereby obtaining a dual-channel hybrid prediction model corresponding to the current control period, and the construction of the dual-channel hybrid prediction model is completed.

[0105] It can be understood that the dual-channel hybrid prediction model combines the piecewise linear regression model with the dynamic compensation mechanism, thereby improving the model accuracy and interpretability. The dynamic compensation auxiliary channel introduces the residual-driven dynamic Q value Kalman filtering algorithm to automatically adapt to changes in working conditions such as device aging and environmental fluctuations. This adaptive dual-channel hybrid prediction model does not require massive training data and significantly reduces the calculation delay, thereby meeting the precise start-up control requirements of high-energy-consumption devices in the workshop and achieving a breakthrough in both algorithm architecture and engineering practicability, which is significantly different from the traditional localized deployment of neural network models and random forest model solutions.

[0106] It can be understood that the dual-channel hybrid prediction model of the embodiment adopts a cloud deployment method, which eliminates the cost of purchasing local servers and the cost of developing desktop software. The YAML files of the prediction model workflow and the bias detection workflow can be quickly exported and copied to other devices for application, thereby shortening the deployment period and having significant advantages in large-scale promotion.

[0107] In a feasible implementation, the step S21 can further include steps C11-C14.

[0108] Step C11, determining the start-up temperature corresponding to each control period, the start-up time corresponding to each control period, the critical temperature rising time corresponding to each control period and the temperature rising standard time corresponding to each control period of the industrial heating device in the historical time period according to the historical temperature rising data of the industrial heating device.

[0109] It should be noted that the control period refers to the time period corresponding to the closed-loop control process of completing one start-up heating, reaching the target device temperature and stopping heating of the industrial heating device, and in the embodiment, the control period is set to be every day. The start-up temperature refers to the initial temperature of the industrial heating device at the moment of starting heating in a certain control period. The start-up time refers to the start-up time of the industrial heating device in a certain control period. The critical temperature rising time refers to the time when the industrial heating device rises to a preset high temperature critical value (for example, 80℃) in a certain control period. The temperature rising standard time refers to the time when the actual temperature of the industrial heating device reaches the target device temperature for the first time from starting heating in a certain control period.

[0110] It can be understood that the historical temperature rising data is analyzed to extract the start-up temperature, start-up time, critical temperature rising time and temperature rising standard time corresponding to each control period.

[0111] Step C12, calculating the temperature rising rate according to the start-up temperature corresponding to each control period, the start-up time corresponding to each control period, the temperature rising standard time corresponding to each control period and the target device temperature corresponding to each control period to obtain the basic temperature rising rate of the industrial heating device.

[0112] It should be noted that for each control period, the temperature rising rate is calculated according to the following formula: temperature rising rate V1=(target device temperature-start-up temperature) / (temperature rising standard time-start-up time), and the average value of the temperature rising rates V1 of all control periods is calculated to obtain the basic temperature rising rate K1 of the industrial heating device.

[0113] Step C13, calculating the temperature rising rate according to the critical temperature rising time corresponding to each control period, the temperature rising standard time corresponding to each control period, the target device temperature corresponding to each control period and the target critical temperature corresponding to each control period to obtain the high temperature rising rate of the industrial heating device.

[0114] It should be noted that for each control period, the temperature rising rate is calculated according to the following formula: temperature rising rate V2=(target device temperature-target critical temperature) / (temperature rising standard time-critical temperature rising time), and the average value of the temperature rising rates V2 of all control periods is calculated to obtain the high temperature rising rate (i.e. high temperature zone temperature rising rate) K2 of the industrial heating device. In the embodiment, the target critical temperature refers to the pre-set high temperature critical value.

[0115] Step C14, obtaining the equipment temperature rising rate according to the basic temperature rising rate and the high temperature rising rate.

[0116] It should be noted that the basic temperature rising rate and the high temperature rising rate are integrated to obtain a comprehensive rate parameter covering the equipment temperature rising process, and the equipment temperature rising rate including the basic temperature rising rate and the high temperature rising rate is obtained.

[0117] In a feasible implementation, the step C12 can further include steps D11-D14.

[0118] Step D11, calculating the temperature difference value corresponding to each control period according to the start temperature corresponding to each control period and the target equipment temperature corresponding to each control period.

[0119] It should be noted that for each control period, the difference between the target equipment temperature and the start temperature is calculated to obtain the temperature difference value corresponding to each control period: target equipment temperature-start temperature.

[0120] Step D12, calculating the time difference value corresponding to each control period according to the start time corresponding to each control period and the temperature rising standard time corresponding to each control period.

[0121] It should be noted that for each control period, the difference between the temperature rising standard time and the start time is calculated to obtain the time difference value corresponding to each control period: temperature rising standard time-start time.

[0122] Step D13, calculating the temperature rising rate corresponding to each control period according to the time difference value corresponding to each control period and the temperature difference value corresponding to each control period.

[0123] Step D14, calculating the mean value of the temperature rising rate corresponding to each control period to obtain the basic temperature rising rate of the industrial heating equipment.

[0124] It should be noted that for each control period, the time difference value and the temperature difference value are used to calculate the temperature rising rate V1 corresponding to each control period, where V1=(target equipment temperature-start temperature) / (temperature rising standard time-start time), and the mean value of the temperature rising rate V1 of all control periods is calculated to obtain the basic temperature rising rate K1 of the industrial heating equipment.

[0125] The embodiment determines a device temperature rising rate by performing temperature rising rate analysis according to historical temperature rising data of the industrial heating device, performs model construction according to the device temperature rising rate, device purging time, target device temperature, preset temperature coefficient, and residual temperature variable, and obtains a main channel regression model, performs compensation deviation analysis on temperature deviation data of the industrial heating device in a historical window according to a dynamic compensation auxiliary channel, determines a dynamic compensation coefficient, and performs deviation compensation on the main channel regression model by using the dynamic compensation coefficient to obtain a double-channel hybrid prediction model. In this way, the model has self-adaptability, and the prediction accuracy of the model is ensured.

[0126] Based on the first embodiment of the present application, in the third embodiment of the present application, the same or similar contents as the above-mentioned first embodiment can be referred to the above description, and will not be described in detail. On this basis, please refer to Figure 5 , step S23, the start-up control method of the industrial heating device further includes steps S01-S03:

[0127] Step S01, according to the dynamic compensation auxiliary channel, the variance of the temperature deviation data of the industrial heating device in the historical window is calculated, and the temperature deviation variance of the industrial heating device is obtained.

[0128] It should be noted that since the temperature deviation data includes the temperature difference value obtained by subtracting the actual device temperature at the target working time from the target device temperature in each control period within the historical window, the temperature difference values of multiple control periods in the temperature deviation data are used for variance calculation, so as to obtain the temperature deviation variance 2 The greater the temperature deviation variance is, the more intense the recent deviation fluctuation of the industrial heating device is, and the process noise covariance Q n needs to be increased to speed up the filtering response speed.

[0129] Step S02, modeling is performed according to the temperature deviation variance, the adaptive coefficient, and the off-diagonal element matrix to obtain a dynamic deviation index.

[0130] It should be noted that the adaptive coefficient a is a parameter for describing the self-adaptive ability of the industrial heating device control system, and its value can be dynamically adjusted according to the system state, and its value range is 0.1-0.2. The off-diagonal element matrix indicates the weak correlation between the long-term deviation and the short-term deviation.

[0131] It can be understood that modeling according to the temperature deviation variance, the adaptive coefficient, and the off-diagonal element matrix can generate the dynamic Q value corresponding to the current control period n, which is In this embodiment, the dynamic deviation index refers to the dynamic Q value corresponding to the current control period n.

[0132] Step S03, determining a dynamic compensation coefficient according to the dynamic deviation index and a Kalman filtering algorithm.

[0133] It should be noted that the dynamic deviation index Q n , a Kalman filtering algorithm is executed, and Kalman filtering iteration is performed to obtain the dynamic compensation coefficient F of the current control period.

[0134] In a feasible implementation, the step S03 can further include steps E11-E15:

[0135] Step E11, constructing a state equation according to the dynamic deviation index and a historical compensation coefficient to obtain a deviation compensation coefficient.

[0136] It should be noted that the historical compensation coefficient refers to the deviation compensation coefficient F n-1 of the previous control period n-1; and the state equation is constructed by combining the dynamic deviation index and the historical compensation coefficient to obtain the deviation compensation coefficient F n of the n-th control period, specifically: F n = F n-1 + W n , wherein W n ~ N(0, Q n ) is a dynamic process noise, and is subject to a multi-dimensional normal distribution with a mean of 0 and a covariance matrix of Q n .

[0137] Step E12, constructing an observation equation according to the deviation compensation coefficient and a target observation noise to obtain a target observation value.

[0138] It should be noted that the target observation noise is a pre-set fixed observation noise R = (0.5℃) 2 , and is a random error generated due to environmental interference and device fluctuation when observing the temperature of the industrial heating device. The state equation is constructed by using the deviation compensation coefficient and the target observation noise to obtain the target observation value Z n of the n-th control period, specifically: Z n = F n + V n , wherein V n ~ N(0, R).

[0139] Step E13, calculating a prediction node according to a historical error covariance and the dynamic deviation index to obtain a prediction error covariance.

[0140] It should be noted that the historical error covariance refers to the error covariance P n-1|n-1 of the previous control period n-1; and the prediction error covariance is calculated by combining the historical error covariance and the dynamic deviation index Q nThe prediction node variance is constructed to obtain the prediction error covariance P of the current control period n|n-1 , specifically P n|n-1 = P n-1|n-1 + Q n . At the same time, the prediction state estimation value of the current control period is constructed, specifically In the embodiment, the initial values of the error covariance and the state estimation value can be determined according to prior knowledge or historical data. For example, if there is a rough estimate of the initial deviation compensation coefficient, the initial value of the state estimation value can be set as the estimate, and the initial value of the error covariance can be set according to the uncertainty of the initial estimate. The greater the uncertainty, the greater the value of the initial value of the error covariance.

[0141] Step E14, Kalman gain calculation is performed according to the prediction error covariance, the target observation noise and the historical error covariance to obtain the Kalman gain.

[0142] Step E15, coefficient calculation is performed according to the target observation value, the Kalman gain and the prediction state estimation value to obtain the dynamic compensation coefficient.

[0143] It should be noted that in the update phase, Kalman gain calculation is performed using the prediction error covariance, the target observation noise and the historical error covariance to obtain the Kalman gain K n , specifically K n = P n-1|n-1 (P n-1|n-1 + R) -1 . Further, the dynamic compensation coefficient F of the current control period n is calculated using the target observation value Z n , the Kalman gain K n , and the prediction state estimation value , specifically And the error covariance P n|n of the optimal estimation value of the deviation compensation coefficient of the current control period n is (I-K n )P n|n-1 .

[0144] It can be understood that in the dynamic compensation auxiliary channel, a Kalman filter value early warning mechanism is set, and when the dynamic compensation coefficient F is greater than the coefficient threshold, a warning message is triggered to prompt the administrator to confirm the model parameters.

[0145] The embodiment calculates the temperature deviation variance of the industrial heating equipment by performing variance calculation on the temperature deviation data of the industrial heating equipment in a history window according to a dynamic compensation auxiliary channel, obtains a temperature deviation variance of the industrial heating equipment, performs modeling according to the temperature deviation variance, an adaptive coefficient and a non-diagonal element matrix, and obtains a dynamic deviation index, and determines a dynamic compensation coefficient according to the dynamic deviation index and a Kalman filtering algorithm. In the foregoing manner, the dynamic compensation coefficient of the current control period can be accurately obtained, the system robustness can be enhanced, and the device state change can be responded in real time.

[0146] It should be noted that the above examples are only used for understanding the present application and do not constitute a limitation on the start-up control method of the industrial heating equipment of the present application. More forms of simple transformation based on the technical concept are within the protection scope of the present application.

[0147] The present application also provides a start-up control device of an industrial heating equipment, which refers to Figure 6 The start-up control device of the industrial heating equipment comprises:

[0148] The acquisition module 10 is configured to acquire the device residual temperature of the industrial heating equipment sent by the Internet of Things platform.

[0149] The prediction module 20 is configured to input the device residual temperature into a double-channel hybrid prediction model, perform time prediction through the double-channel hybrid prediction model, and obtain a temperature rise prediction duration of the industrial heating equipment. The double-channel hybrid prediction model is obtained by compensating a main channel regression model through a dynamic compensation auxiliary channel.

[0150] The control module 30 is configured to perform start-up control on the industrial heating equipment according to the temperature rise prediction duration.

[0151] Optionally, the prediction module 20 is further configured to:

[0152] perform temperature rise rate analysis according to historical temperature rise data of the industrial heating equipment, determine a device temperature rise rate, perform model construction according to the device temperature rise rate, a device purging time, a target device temperature, a preset temperature coefficient and a residual temperature variable, obtain a main channel regression model, perform compensation deviation analysis on temperature deviation data of the industrial heating equipment in a history window according to a dynamic compensation auxiliary channel, determine a dynamic compensation coefficient, and perform deviation compensation on the main channel regression model through the dynamic compensation coefficient to obtain a double-channel hybrid prediction model.

[0153] Optionally, the prediction module 20 is further configured to:

[0154] The starting temperature corresponding to each control period, the starting time corresponding to each control period, the temperature rising critical time corresponding to each control period, and the temperature rising standard time corresponding to each control period of the industrial heating equipment in a historical time period are determined according to historical temperature rising data of the industrial heating equipment; the basic temperature rising rate of the industrial heating equipment is calculated according to the starting temperature corresponding to each control period, the starting time corresponding to each control period, the temperature rising standard time corresponding to each control period, and the target equipment temperature corresponding to each control period; the high temperature rising rate of the industrial heating equipment is calculated according to the temperature rising critical time corresponding to each control period, the temperature rising standard time corresponding to each control period, the target equipment temperature corresponding to each control period, and the target critical temperature corresponding to each control period; and the equipment temperature rising rate is obtained according to the basic temperature rising rate and the high temperature rising rate.

[0155] Optionally, the prediction module 20 is further configured to:

[0156] The temperature difference corresponding to each control period is calculated according to the starting temperature corresponding to each control period and the target equipment temperature corresponding to each control period, to obtain a temperature difference value corresponding to each control period; the time difference corresponding to each control period is calculated according to the starting time corresponding to each control period and the temperature rising standard time corresponding to each control period, to obtain a time difference value corresponding to each control period; the temperature rising rate corresponding to each control period is calculated according to the time difference value corresponding to each control period and the temperature difference value corresponding to each control period; and the mean value of the temperature rising rate corresponding to each control period is calculated to obtain the basic temperature rising rate of the industrial heating equipment.

[0157] Optionally, the prediction module 20 is further configured to:

[0158] The temperature deviation variance of the industrial heating equipment in a historical window is calculated according to the dynamic compensation auxiliary channel, to obtain the temperature deviation variance of the industrial heating equipment; the dynamic deviation index is obtained by modeling according to the temperature deviation variance, an adaptive coefficient, and a non-diagonal element matrix; and the dynamic compensation coefficient is determined according to the dynamic deviation index and a Kalman filtering algorithm.

[0159] Optionally, the prediction module 20 is further configured to:

[0160] According to the dynamic deviation index and the historical compensation coefficient, a state equation is constructed to obtain a deviation compensation coefficient; according to the deviation compensation coefficient and a target observation noise, an observation equation is constructed to obtain a target observation value; according to a historical error covariance and the dynamic deviation index, a prediction node is calculated to obtain a prediction error covariance; according to the prediction error covariance, the target observation noise and the historical error covariance, a Kalman gain is calculated to obtain a Kalman gain; and according to the target observation value, the Kalman gain and a prediction state estimation value, a coefficient is calculated to obtain a dynamic compensation coefficient.

[0161] Optionally, the control module 30 is further configured to:

[0162] According to the target working time and the temperature rise prediction duration, a predicted start-up time of the industrial heating equipment is determined; and the predicted start-up time is pushed to a terminal of an administrator.

[0163] Optionally, the control module 30 is further configured to:

[0164] An actual equipment temperature of the industrial heating equipment at the target working time is obtained; an actual temperature deviation between the target equipment temperature and the actual equipment temperature is calculated; and when the actual temperature deviation is not within a preset deviation range, temperature warning information is pushed to a terminal of an administrator for deviation warning.

[0165] The industrial heating equipment start-up control device provided in the application adopts the industrial heating equipment start-up control method in the above embodiments, and can solve the technical problems of the preheating control scheme of the high-energy-consumption equipment in the prior art, such as serious energy waste, poor interpretability of the prediction model and large cumulative error. Compared with the prior art, the industrial heating equipment start-up control device provided in the application has the same beneficial effects as the industrial heating equipment start-up control method provided in the above embodiments, and other technical features in the industrial heating equipment start-up control device are the same as the features disclosed in the above embodiments, and thus will not be described here.

[0166] The application provides a cloud server, which comprises at least one processor and a memory in communication connection with the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the industrial heating equipment start-up control method in the above embodiment one.

[0167] Reference will be made to the following description of embodiments of the application Figure 7, which shows a schematic diagram of the structure of a cloud server suitable for implementing the embodiments of the present application. The cloud server in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 7 The cloud server shown is merely an example and should not limit the functionality and scope of use of the embodiments of the present application.

[0168] like Figure 7 As shown, the cloud server may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. Various programs and data required for the operation of the cloud server are also stored in RAM 1004. The processing device 1001, ROM 1002, and RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the cloud server to communicate with other cloud servers wirelessly or wired to exchange data. Although the figure shows a cloud server with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

[0169] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program codes for executing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiments disclosed in the present application are executed.

[0170] The cloud server provided by the present application adopts the start-up control method of the industrial heating equipment in the above-mentioned embodiments, which can solve the technical problems of serious energy waste, poor interpretability of the prediction model and large cumulative error in the preheating control scheme of high-energy-consumption equipment in the prior art. Compared with the prior art, the cloud server provided by the present application has the same beneficial effects as the start-up control method of the industrial heating equipment provided by the above-mentioned embodiments, and other technical features in the cloud server are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0171] It should be understood that parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the description of the above-mentioned embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0172] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0173] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e. computer program) for executing the start-up control method of the industrial heating equipment in the above-mentioned embodiments.

[0174] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, 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 thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0175] The computer-readable storage medium may be included in the cloud server, or may exist independently without being installed in the cloud server.

[0176] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the cloud server, the cloud server: obtains the equipment residual temperature of the industrial heating equipment sent by the Internet of Things platform; inputs the equipment residual temperature into the dual-channel hybrid prediction model, performs time prediction through the dual-channel hybrid prediction model, and obtains the predicted heating time of the industrial heating equipment. The dual-channel hybrid prediction model is obtained by compensating the main channel regression model through the dynamic compensation auxiliary channel; and controls the power on of the industrial heating equipment according to the predicted heating time.

[0177] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0178] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0179] The modules involved in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0180] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the startup control method of the industrial heating equipment, and can solve the technical problems of the preheating control scheme of the high-energy-consumption equipment in the prior art, such as serious energy waste, poor interpretability of the prediction model, and large cumulative error. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the industrial heating equipment startup control method provided by the above embodiments, and will not be described here.

[0181] The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the start-up control method of the industrial heating device as described above.

[0182] The computer program product provided by the application can solve the technical problems of the preheating control scheme of the high-energy-consumption device in the prior art, such as serious energy waste, poor interpretability of the prediction model, and large cumulative error. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the start-up control method of the industrial heating device provided by the above-mentioned embodiments, and will not be repeated here.

[0183] The above only describes some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation made by using the content of the specification and drawings, or direct / indirect application in other related technical fields under the technical concept of the application is included in the patent protection scope of the application.

Claims

1. A startup control method for industrial heating equipment, characterized in that: The method is applied to a cloud server, which is in communication with an Internet of Things platform, and includes: Obtain the residual temperature of industrial heating equipment sent by the IoT platform; The residual temperature of the equipment is input into a dual-channel hybrid prediction model, and time prediction is performed through the dual-channel hybrid prediction model to obtain the predicted temperature rise time of the industrial heating equipment. The dual-channel hybrid prediction model is obtained by compensating the main channel regression model through the dynamic compensation auxiliary channel; The industrial heating equipment is controlled to start up according to the predicted temperature rise time.

2. The method according to claim 1, wherein Before the step of inputting the residual temperature of the equipment into a dual-channel hybrid prediction model and performing time prediction by the dual-channel hybrid prediction model to obtain the predicted heating time of the industrial heating equipment, the method further includes: Performing a heating rate analysis based on historical heating data of the industrial heating equipment to determine the heating rate of the equipment; A model is constructed based on the device heating rate, device purge time, target device temperature, preset temperature coefficient, and residual temperature variable to obtain a main channel regression model; Perform compensation deviation analysis on the temperature deviation data of industrial heating equipment within the historical window based on the dynamic compensation auxiliary channel to determine the dynamic compensation coefficient; The main channel regression model is compensated for deviations using the dynamic compensation coefficient to obtain a dual-channel hybrid prediction model.

3. The method according to claim 2, wherein The step of analyzing the heating rate according to the historical heating data of the industrial heating equipment to determine the heating rate of the equipment includes: Determining, based on historical temperature rise data of the industrial heating equipment, the starting temperature corresponding to each control period, the starting time corresponding to each control period, the temperature rise critical time corresponding to each control period, and the temperature rise target time corresponding to each control period of the industrial heating equipment within a historical time period; The heating rate is calculated based on the starting temperature corresponding to each control cycle, the starting time corresponding to each control cycle, the temperature reaching standard time corresponding to each control cycle, and the target device temperature corresponding to each control cycle to obtain the basic heating rate of the industrial heating equipment; The heating rate is calculated based on the critical heating time corresponding to each control cycle, the heating target time corresponding to each control cycle, the target equipment temperature corresponding to each control cycle, and the target critical temperature corresponding to each control cycle to obtain the high-temperature heating rate of the industrial heating equipment; The equipment heating rate is obtained according to the basic heating rate and the high temperature heating rate.

4. The method according to claim 3, wherein The step of calculating the heating rate according to the starting temperature corresponding to each control cycle, the starting time corresponding to each control cycle, the temperature reaching target time corresponding to each control cycle, and the target device temperature corresponding to each control cycle to obtain the basic heating rate of the industrial heating equipment includes: The temperature difference is calculated based on the starting temperature corresponding to each control cycle and the target device temperature corresponding to each control cycle to obtain the temperature difference value corresponding to each control cycle; The time difference is calculated based on the start time corresponding to each control cycle and the temperature rise reaching standard time corresponding to each control cycle to obtain the time difference value corresponding to each control cycle; The rate calculation is performed based on the time difference corresponding to each control period and the temperature difference corresponding to each control period to obtain the heating rate corresponding to each control period; The heating rates corresponding to each control period are averaged to obtain the basic heating rate of the industrial heating equipment.

5. The method according to claim 2, wherein The step of performing compensation deviation analysis on the temperature deviation data of the industrial heating equipment within the historical window according to the dynamic compensation auxiliary channel to determine the dynamic compensation coefficient includes: performing variance calculation on the temperature deviation data of the industrial heating equipment within a historical window according to the dynamic compensation auxiliary channel to obtain the temperature deviation variance of the industrial heating equipment; Modeling is performed based on the temperature deviation variance, the adaptive coefficient, and the off-diagonal element matrix to obtain a dynamic deviation index; A dynamic compensation coefficient is determined according to the dynamic deviation index and the Kalman filter algorithm.

6. The method according to claim 5, wherein The step of determining the dynamic compensation coefficient according to the dynamic deviation index and the Kalman filter algorithm includes: A state equation is constructed according to the dynamic deviation index and the historical compensation coefficient to obtain the deviation compensation coefficient; An observation equation is constructed according to the deviation compensation coefficient and the target observation noise to obtain a target observation value; Perform prediction node calculation based on the historical error covariance and the dynamic deviation index to obtain the prediction error covariance; Performing a Kalman gain calculation based on the prediction error covariance, the target observation noise, and the historical error covariance to obtain a Kalman gain; A coefficient is calculated based on the target observation value, the Kalman gain and the predicted state estimate to obtain a dynamic compensation coefficient.

7. The method according to any one of claims 1 to 6, characterized in that The step of controlling the startup of the industrial heating equipment according to the predicted temperature rise time includes: Determining a predicted startup time of the industrial heating equipment according to the target working time and the predicted temperature rise time; Push the predicted boot time to the administrator's terminal.

8. The method according to any one of claims 1 to 6, characterized in that After the step of controlling the startup of the industrial heating equipment according to the predicted temperature rise time, the method further includes: Obtaining the actual device temperature of the industrial heating equipment during the target working time; calculating an actual temperature deviation between a target device temperature and the actual device temperature; When the actual temperature deviation is not within the preset deviation range, the temperature warning information is pushed to the administrator's terminal for deviation warning.

9. A startup control device for industrial heating equipment, characterized in that: The device comprises: The acquisition module is used to obtain the residual temperature of industrial heating equipment sent by the IoT platform; a prediction module, configured to input the residual temperature of the equipment into a dual-channel hybrid prediction model, perform time prediction using the dual-channel hybrid prediction model, and obtain a predicted temperature rise duration of the industrial heating equipment, wherein the dual-channel hybrid prediction model is obtained by compensating the main channel regression model through a dynamic compensation auxiliary channel; A control module is used to control the startup of the industrial heating equipment according to the predicted temperature rise time.

10. A cloud server, characterized in that: The cloud server includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the startup control method for industrial heating equipment according to any one of claims 1 to 8.

11. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the startup control method of the industrial heating equipment according to any one of claims 1 to 8 are implemented.

Images