Prefabricated T-beam intelligent temperature control maintenance system

Abstract

The invention relates to the technical field of prefabricated T-beam maintenance, in particular to an intelligent temperature control maintenance system for a prefabricated T-beam. Comprising a steam curing chamber, a steam generator, a spraying system, an outdoor environment acquisition module, an indoor environment acquisition module, a steam generator parameter acquisition module, a spraying system parameter acquisition module, a prefabricated T-beam temperature acquisition module, a control module, a cooling water control valve and a steam control valve. The control module is used for automatically adjusting the steam amount or the water spraying amount of the steam generator based on preset curing parameters and real-time monitoring data, precise control over the whole steam curing process (standing, heating, constant temperature and cooling) of the prefabricated T-shaped beam is achieved, man-made influences are reduced, the steam curing temperature, humidity and curing time of the prefabricated beam are effectively controlled, the curing time is shortened, and the production efficiency is improved. And the maintenance quality of the prefabricated T-beam is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of precast T-beam maintenance, in particular to a precast T-beam intelligent temperature control maintenance system. BACKGROUND

[0002] With the rapid development of the construction industry in China, precast concrete components and their steam curing process are entering a critical period of development. As a common precast concrete component, precast T-beams are widely used in bridge construction. Traditional precast beam yards have problems such as long curing time, large land occupation, and susceptibility to natural environmental influences. Currently, precast T-beam steam curing usually adopts natural curing and steam curing, of which steam curing can shorten the curing time and improve the formwork turnover rate, but still has some problems.

[0003] Firstly, traditional steam curing equipment can only achieve simple control of the supply steam temperature and cannot respond to changes in outdoor temperature, wind, and other factors, resulting in changes in the steam temperature inside the shed and affecting the curing effect. Secondly, the existing precast concrete component steam curing method has problems such as insufficient heat exchange intensity, high energy consumption, and excessively long curing time. In addition, during the curing process of precast T-beams, heat is generated inside, which can easily cause the internal temperature to be too high, affecting the overall strength and durability.

[0004] Chinese patent application CN118081961A discloses a precast box beam steam curing construction device and its construction process, which includes a curing box with an intelligent temperature and humidity control system inside, and also includes a main steam pipeline, a main spraying pipeline, and a plurality of branch steam pipelines and branch spraying pipelines. The device connects the branch steam pipelines and branch spraying pipelines through the main steam pipeline and the main spraying pipeline, and a plurality of equally spaced steam nozzles are provided on each branch steam pipeline. This device solves the problem of long time for improving the strength and elastic modulus of the box beam during outdoor natural curing, and can quickly reach the tensioning condition after demolding. However, the temperature and humidity control precision and stability of the intelligent temperature and humidity control system of this device still need to be improved.

[0005] In order to solve the above problems, an intelligent system capable of accurately controlling the temperature and humidity of precast T-beam steam curing is urgently needed. This system should be able to automatically adjust the amount of steam or water according to the specific conditions and environmental conditions of the precast T-beam, to achieve targeted care and maintenance of the precast T-beam, thereby reducing the curing time and improving the curing quality. SUMMARY

[0006] In view of the problems in the prior art, the present application provides a precast T-beam intelligent temperature control maintenance system, and the specific technical solutions are as follows: The application discloses a precast T-beam intelligent temperature control maintenance system, which comprises a steam curing chamber, a steam generator, a spraying system, an outdoor environment acquisition module, an indoor environment acquisition module, a steam generator parameter acquisition module, a spraying system parameter acquisition module, a precast T-beam temperature acquisition module, a control module, a cooling water control valve and a steam control valve. The outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module, the precast T-beam temperature acquisition module, the cooling water control valve and the steam control valve are connected with the control module; the steam control valve is connected with the steam generator, and the cooling water control valve is connected with the spraying system. The outdoor environment acquisition module is used for acquiring the temperature and the wind speed of the external environment of the steam curing chamber. The indoor environment acquisition module is used for acquiring the temperature and the humidity of the internal environment of the steam curing chamber. The steam generator parameter acquisition module is used for acquiring the steam flow of the steam generator. The spraying system parameter acquisition module is used for acquiring the cooling water flow of the spraying system. The precast T-beam temperature acquisition module is used for acquiring the temperature inside the precast T-beam. The control module is used for controlling the opening degrees of the cooling water control valve and the steam control valve according to the parameters acquired by the outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module and the precast T-beam temperature acquisition module, so as to control the steam flow of the steam generator and the cooling water flow of the spraying system in each maintenance stage of the precast T-beam maintenance; the maintenance stage comprises four stages of static stop, temperature rise, constant temperature and temperature drop.

[0007] Preferably, the indoor environment acquisition module comprises a plurality of temperature sensors arranged at respective indoor temperature monitoring points inside the steam curing chamber, and the temperature difference between the respective indoor temperature monitoring points is not greater than a preset temperature difference value.

[0008] Preferably, a safety valve is further arranged on the steam curing chamber and connected with the control module, and the indoor environment acquisition module is further used for acquiring the pressure inside the steam curing chamber. When the pressure inside the steam curing chamber is greater than a preset threshold value, the control module controls the safety valve to release the pressure of the steam curing chamber.

[0009] Preferably, the control module is provided with a temperature prediction unit and a temperature control unit; the temperature prediction unit is used for predicting the temperature inside the precast T-beam at the next moment according to the parameters acquired by the outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module and the precast T-beam temperature acquisition module; and the temperature control unit is used for generating the opening degree parameters of the cooling water control valve and the steam control valve at the next moment according to the temperature inside the precast T-beam at the next moment, so as to control the steam flow of the steam generator and the cooling water flow of the spraying system.

[0010] Preferably, the temperature prediction unit comprises an LSTM prediction network, which comprises an input layer, 2 hidden layers, and an output layer, wherein the input feature vector of the input layer is a data sequence collected at time t-1, including the temperature of the environment outside the curing chamber, the wind speed at time t-1, the temperature and humidity of the environment inside the curing chamber at time t-1, the steam flow of the steam generator at time t-1, the cooling water flow of the spraying system at time t-1, and the temperature inside the prefabricated T-beam at time t-1. The hidden layer obtains the hidden state at time t according to the input feature vector of the input layer at time t-1. The output layer obtains the temperature inside the prefabricated T-beam at time t according to the hidden state at time t.

[0011] Preferably, the system introduces an attention mechanism in the LSTM prediction network, calculates an attention context vector according to the attention mechanism, and fuses the attention context vector and the hidden state at time t through vector splicing and a fully connected layer to form a state feature at time t.

[0012] Preferably, the specific steps of introducing the attention mechanism in the LSTM prediction network to form the state feature at time t are as follows: (1) Calculate the attention score of the historical time i to the current time t as follows: ; wherein, is the attention vector, T is the transpose, is the hidden state at time i is the weight of the hidden state at time i is the hidden state at time t is the weight of the hidden state at time t is the bias of the attention, and the tanh function outputs [-1, 1]; (2) Normalize the calculated attention score of the historical time i to the current time t , and the normalized attention score is as follows: ; wherein, , and ; (3) Calculate the attention context vector according to the normalized attention score , which is specifically as follows: ; (4) Fuse the hidden state at time t output by the LSTM prediction network with the attention context vector​ The state feature at time t is formed by vector splicing and full connection layer fusion , specifically as follows: ; Wherein, is the target deviation feature vector at time t, , is the fusion weight and bias, including the temperature deviation at time t, the temperature change rate deviation at time t, and the humidity deviation at time t.

[0013] Preferably, the temperature control unit comprises a DQP decision network, which calculates state-action value Q by setting state space, action space and reward function, and obtains the optimal action space according to the calculated state-action value Q, the action space including the valve opening of the steam control valve and the valve opening of the cooling water control valve.

[0014] Preferably, the step of obtaining the optimal action space by the DQP decision network is as follows: (1) Define the state space of the DQP decision network is: ; Wherein is the stage identifier, , respectively representing the static stop, temperature rise, constant temperature, and temperature drop stage; The action space of the DQP decision network is defined as: ; Wherein, is the valve opening of the steam control valve at time t, is the valve opening of the cooling water control valve at time t; (2) The calculation of state-action value Q value is as follows: ; Wherein, is the parameter of the DQP decision network, FC is the full connection layer, and Q value represents the future cumulative reward when action is taken in state ; (3) The reward function is set as follows: ; , , is the weight, and + + =1, and , is a temperature deviation at time t, is a temperature change rate deviation at time t, is a humidity deviation at time t; (4) using time series difference error to optimize network parameters , and selecting the current maximum action space as the optimal action space with a probability of 1-epsilon.

[0015] Preferably, the prefabricated T-beam temperature acquisition module is further used to acquire the temperature outside the prefabricated T-beam, and the control module is further used to control the temperature outside the prefabricated T-beam to be less than the temperature inside the prefabricated T-beam by a preset temperature difference.

[0016] Compared with the prior art, the present application has the following beneficial effects: The present application automatically adjusts the steam amount or water spraying amount of the steam generator based on the preset curing parameters and real-time monitoring data through the control module, realizes precise control of the whole process (static stop, temperature rise, constant temperature, temperature drop) of the prefabricated T-beam steam curing, reduces human influence, effectively controls the curing temperature, humidity and curing time of the prefabricated beam, reduces the curing time, and ensures the curing quality of the prefabricated T-beam. The present application considers the influence of the parameters of the internal and external environment of the curing chamber on the internal temperature of the prefabricated T-beam, which is beneficial to improve the control precision of the internal temperature of the prefabricated T-beam. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the specific embodiments or prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale.

[0018] Fig. 1 is a system block diagram of the present application.

[0019] Fig. 2 is a working flowchart of the system of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] It should be understood that the terms "comprises" and "comprising," when used in this specification and accompanying claims, indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0022] It should also be understood that the terms used in the specification and the appended claims are intended to be interpreted broadly and in a manner consistent with the principles of the prior art, and that references to a particular embodiment or a particular use of the application are intended to refer only to that specific embodiment or use, and that the use of the terms "comprises," "comprising," "includes," "including" and the like are not intended to exclude the presence or addition of one or more other features, integers, steps, operations, elements, and / or groups thereof.

[0023] It should further be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items, and that the term "at least one of' denotes one, or a plurality of, or any combination of the listed items.

[0024] As shown in the Figs. 1-2 The embodiment provides a prefabricated T-beam intelligent temperature control maintenance system, which comprises a steam curing chamber, a steam generator, a spraying system, an outdoor environment acquisition module, an indoor environment acquisition module, a steam generator parameter acquisition module, a spraying system parameter acquisition module, a prefabricated T-beam temperature acquisition module, a control module, a cooling water control valve and a steam control valve.

[0025] The outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module, the prefabricated T-beam temperature acquisition module, the cooling water control valve and the steam control valve are connected with the control module; the steam control valve is connected with the steam generator, and the cooling water control valve is connected with the spraying system.

[0026] The outdoor environment acquisition module is used for acquiring the temperature and the wind speed of the external environment of the steam curing chamber; the indoor environment acquisition module is used for acquiring the temperature and the humidity of the internal environment of the steam curing chamber; the steam generator parameter acquisition module is used for acquiring the steam flow of the steam generator; the spraying system parameter acquisition module is used for acquiring the cooling water flow of the spraying system; and the prefabricated T-beam temperature acquisition module is used for acquiring the temperature inside the prefabricated T-beam.

[0027] The control module is used for controlling the opening degrees of the cooling water control valve and the steam control valve according to the parameters acquired by the outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module and the prefabricated T-beam temperature acquisition module, so as to control the steam flow of the steam generator and the cooling water flow of the spraying system in each maintenance stage of the prefabricated T-beam maintenance; the maintenance stage comprises four stages of static stop, temperature rise, constant temperature and temperature drop.

[0028] As a preferred specific embodiment, the indoor environment acquisition module acquires the temperature of the interior environment of the steam curing chamber by arranging multiple temperature sensors at multiple indoor temperature monitoring points in the interior of the steam curing chamber, and the temperature difference between the multiple indoor temperature monitoring points is not greater than a temperature difference preset value, which is set to 15°C in this embodiment. In this way, the temperature of each part of the interior of the steam curing chamber can be made as consistent as possible.

[0029] As a preferred specific embodiment, a safety valve is further arranged on the steam curing chamber and connected to the control module, and the indoor environment acquisition module is further configured to acquire the pressure in the interior of the steam curing chamber. When the pressure in the interior of the steam curing chamber is greater than a preset threshold, the control module controls the safety valve to release the pressure of the steam curing chamber.

[0030] As a preferred specific embodiment, the control module is provided with a temperature prediction unit and a temperature control unit. The temperature prediction unit is configured to predict the temperature in the precast T beam at the next moment according to the parameters acquired by the outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module, and the precast T beam temperature acquisition module. The temperature control unit is configured to generate the opening parameter of the cooling water control valve and the steam control valve at the next moment according to the temperature in the precast T beam at the next moment, so as to control the steam flow of the steam generator and the cooling water flow of the spraying system.

[0031] As a preferred specific embodiment, the temperature prediction unit includes an LSTM prediction network, which includes an input layer, two hidden layers, and an output layer. The input feature vector of the input layer is a data sequence acquired at t-1 moment, including the temperature and wind speed of the environment outside the steam curing chamber at t-1 moment, the temperature and humidity of the environment inside the steam curing chamber at t-1 moment, the steam flow of the steam generator at t-1 moment, the cooling water flow of the spraying system at t-1 moment, and the temperature inside the precast T beam at t-1 moment. The hidden layer obtains the hidden state at t moment according to the input feature vector of the input layer at t-1 moment. The output layer obtains the temperature inside the precast T beam at t moment according to the hidden state at t moment.

[0032] In this embodiment, the input feature vector is further preprocessed before being input into the input layer of the LSTM prediction network, specifically including the following steps: (1) The 3σ criterion is used to eliminate abnormal values, and the formula is: When or , the corresponding data is eliminated. is any parameter in the input feature vector at t-1 moment, and the mean of the previous and subsequent two data of the parameter is used to correct the original , i.e.: ; The modified .

[0033] (2) The input feature vector is mapped to the interval [0, 1], and the formula is: ; Wherein, The minimum value of the element , The maximum value of the element , The Normalized value.

[0034] The calculation formula of each gate of the LSTM prediction network is as follows: (1) The forgetting gate is to decide to discard the cell state Information, the calculation formula is as follows: Wherein, The output of the forgetting gate at t time, The weight of the forgetting gate, The bias of the forgetting gate, The hidden state at t-1 time, That is The normalized value , Sigmoid activation function, output [0, 1]; (2) The input gate determines the new information, and the calculation formula is: ; ; The calculation method of the updated amount Of the cell state is as follows: ; Wherein, The output of the input gate, , The weight and bias of the input gate, The cell state update amount, , The weight and bias of the cell state update; tanh function output [-1, 1]; The element multiplication is.

[0035] (3) The output gate generates the current hidden state, and the specific calculation formula is as follows: wherein, is the output of the output gate at time t, , are the weight and bias of the output gate, respectively; is the hidden state at time t. The final LSTM prediction network outputs a sequence of hidden states , reflecting the time series characteristics.

[0036] The output layer maps the hidden state at time t to the predicted beam internal temperature value at time t through a fully connected layer. ; wherein, , are the weight and bias of the output layer, respectively.

[0037] In the training and optimization process of the LSTM prediction network, the Adam optimizer is used, the learning rate is 0.001, and the number of iterations is set to 100.

[0038] The loss function is as follows: ; wherein, is the i-th predicted temperature value, is the actual i-th temperature value, and m is the number of data samples. When is less than or equal to a set value, the LSTM prediction network converges.

[0039] As a preferred specific embodiment, the system introduces an attention mechanism in the LSTM prediction network, which is used to highlight the historical time that has the greatest impact on the current maintenance stage (such as the temperature 30 minutes before the constant temperature stage is more critical for the current adjustment), calculates an attention context vector according to the attention mechanism, and fuses the attention context vector and the hidden state at time t through vector splicing and a fully connected layer to form a state feature at time t.

[0040] As a preferred specific embodiment, the attention mechanism is introduced in the LSTM prediction network to form the specific steps of the state feature at time t as follows: (1) For the LSTM prediction network output hidden state sequence , the attention score of historical time i to the current time t is calculated as follows: ; wherein, is the attention vector, T is the transpose, is the weight of the hidden state at time i, is the hidden state at time t The weight, As a bias for attention, the tanh function outputs [-1, 1].

[0041] (2) Calculate the attention score of historical time i to the current time t. Normalization is performed, and the normalized attention score is obtained. as follows: ; in, ,and ; This indicates the importance of historical moment i.

[0042] (3) Based on the normalized attention score Calculate attention context vector The details are as follows: ; Attention Context Vector This is a weighted sum of historical hidden states, highlighting the historical moments that have the greatest impact on the current state, for example: Thermostatic phase: Temperature fluctuations in the first 10 minutes (affecting whether the steam volume needs to be fine-tuned); Cooling phase: The cooling rate in the first 20 minutes (determines whether the current cooling water flow rate needs to be increased); Special operating conditions: such as when the ambient humidity drops suddenly, the effective experience of historical humidity regulation.

[0043] (4) Output the hidden state at time t from the LSTM prediction network. With attention context vector The state features at time t are formed by vector concatenation and fusion through a fully connected layer. The details are as follows: ; in, Let be the target deviation feature vector at time t. , To integrate weights and biases, optimization is achieved through training.

[0044] This includes the temperature deviation at time t, the temperature change rate deviation at time t, and the humidity deviation at time t. The calculation method is as follows: Temperature deviation at time t: ; Deviation in the rate of temperature change at time t: ; ; humidity deviation at time t: ; i.e. ; wherein, target temperature of the target curve, represents the difference from the target curve; standard rate of temperature rise or fall, represents the temperature change amount of ; represents the difference between the actual temperature rise / fall rate and the standard rate; target humidity, represents the difference from the target humidity.

[0045] As a preferred embodiment, the temperature control unit comprises a DQP decision network, which calculates state-action value Q by setting state space, action space and reward function, and obtains the optimal action space according to the calculated state-action value Q, the action space including the valve opening of the steam control valve and the valve opening of the cooling water control valve.

[0046] As a preferred embodiment, the step of obtaining the optimal action space by the DQP decision network is as follows: (1) Define the state space of the DQP decision network as: ; wherein is the stage identifier, , respectively, represent the static stop, temperature rise, constant temperature, and temperature drop stages; compressing real-time monitoring data, historical time series rules, key influencing factors, and target deviation into a feature vector that can be directly processed by the DQP decision network, so that the DQP decision network can output the optimal steam / cooling water flow adjustment strategy based on comprehensive information, and ensure stable tracking of the target curve during maintenance.

[0047] Define the action space of the DQP decision network as: ; wherein, is the valve opening of the steam control valve at time t, is the valve opening of the cooling water control valve at time t.

[0048] (2) The calculation of the state-action value Q value is as follows: ; wherein, is the parameter of the DQP decision network, FC is the full connection layer, and Q value represents the state Take action Future cumulative reward.

[0049] (3) The reward function is set as follows: ; , , is a weight, and + + =1, and , is the temperature deviation at time t, is the temperature change rate deviation at time t, is the humidity deviation at time t.

[0050] (4) The time series difference TD error is used to optimize the network parameters , and the current maximum action space is selected as the optimal action space with a probability of 1-epsilon. The time series difference TD error is used to optimize the network parameters Specifically as follows: ; Wherein, is a discount factor, indicating the weight of future rewards, is the target parameter of the network, which is copied from theta regularly to improve stability.

[0051] The current maximum action is selected with a probability of 1-epsilon, and the action is randomly selected with a probability of epsilon, and epsilon decays with training iterations (such as from 0.9 to 0.1).

[0052] After the DQP decision network obtains the optimal action space, the steam flow and cooling water flow are controlled according to the action space. After the action space acts on the system, the system reacquires parameters to form a closed loop control.

[0053] The system captures time series dependence through the LSTM prediction network and adds an attention mechanism to focus on key history, and then learns the optimal action space through the DQP decision network to realize dynamic adjustment of steam / cooling water flow. The core is to approximate the ideal maintenance curve in a data-driven manner. In practical applications, the parameter range and weight coefficient are adjusted in combination with the prefabricated T-beam maintenance specification (such as JT / T722-2023).

[0054] In the embodiment, the steam curing chamber contains 5 independent heat insulation fully enclosed sheds, the single shed is 65m long, 3.5m high and 3m wide, and 10 prefabricated T-beams can be cured simultaneously. The steam curing chamber adopts the design of independent heat insulation fully enclosed shed, which is composed of light steel structure and thermal insulation cotton. The steam curing chamber is more easily sealed, the heat loss is small, and the cost is more easily controlled.

[0055] As a preferred specific embodiment, the prefabricated T-beam temperature acquisition module is also used to acquire the temperature outside the prefabricated T-beam, and the control module is also used to control the temperature outside the prefabricated T-beam to be less than the preset temperature difference value from the temperature inside the prefabricated T-beam.

[0056] As a preferred specific embodiment, in the embodiment, whether the steam curing chamber is good in air tightness is judged by the pressure of the internal environment of the steam curing chamber in the static stopping stage, that is, the pressure difference of adjacent moments is less than a preset value, which is good in air tightness, and only when the steam curing chamber is good in air tightness, the temperature rising stage is entered.

[0057] The temperature rising speed of the temperature rising stage is not greater than 10℃ / h, the constant temperature temperature is controlled at about 45℃, and the time of constant temperature curing is determined through tests according to the demoulding strength requirement of the prefabricated T-beam, the concrete mix proportion condition and the specific environmental condition.

[0058] The internal temperature of the prefabricated T-beam should not be higher than 60℃ during the constant temperature steam curing, which is monitored by the temperature sensor buried in the prefabricated T-beam, when the internal temperature of the prefabricated T-beam is too high, the method of reducing or stopping steam supply to the steam curing chamber is adopted to reduce the temperature in the shed, and the temperature difference between the inside and outside of the prefabricated T-beam should be not greater than a preset value, which is 15℃ in the embodiment. The temperature falling speed of the temperature falling stage should not be greater than 10℃ / h.

[0059] Those skilled in the art can understand that the modules of the examples described in combination with the embodiments disclosed in the present text can be realized in electronic hardware, computer software or combination of both, in order to clearly illustrate the interchangeability of hardware and software, the composition of the examples has been described in general according to functions in the above description. Whether the functions are realized in hardware or software mode depends on the specific application and design constraints of the technical solution. The professional technicians can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0060] In the embodiments provided in the present application, it should be understood that the division of modules is only a logical functional division, and when actually implemented, there can be another division mode, for example, multiple modules can be combined into one module, one module can be split into multiple modules, or some features can be ignored, etc.

[0061] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically independently, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module.

[0062] When the integrated module is realized in the form of a software functional module and sold or used as an independent product, the integrated module can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0063] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the specification of the present application.

Claims

1. A precast T-beam intelligent temperature control maintenance system, characterized in that, The steam curing chamber comprises a steam curing chamber, a steam generator, a spraying system, an outdoor environment acquisition module, an indoor environment acquisition module, a steam generator parameter acquisition module, a spraying system parameter acquisition module, a prefabricated T-beam temperature acquisition module, a control module, a cooling water control valve, and a steam control valve. The outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module, the prefabricated T-beam temperature acquisition module, the cooling water control valve, and the steam control valve are connected to the control module. The outdoor environment acquisition module is configured to acquire the temperature and wind speed of the environment outside the steam curing chamber. The indoor environment acquisition module is configured to acquire the temperature and humidity of the environment inside the steam curing chamber. The steam generator parameter acquisition module is configured to acquire the steam flow of the steam generator. The spraying system parameter acquisition module is configured to acquire the cooling water flow of the spraying system. The prefabricated T-beam temperature acquisition module is configured to acquire the temperature inside the prefabricated T-beam. The control module is configured to control the opening degree of the cooling water control valve and the steam control valve according to the parameters acquired by the outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module, and the prefabricated T-beam temperature acquisition module, and to control the steam flow of the steam generator and the cooling water flow of the spraying system in each curing stage of the prefabricated T-beam curing.

2. The precast T-beam intelligent temperature control maintenance system according to claim 1, characterized in that, The indoor environment acquisition module comprises a plurality of temperature sensors arranged at respective indoor temperature monitoring points inside the steam curing chamber, and the temperature difference between any two of the indoor temperature monitoring points is not greater than a preset temperature difference value.

3. The precast T-beam intelligent temperature control maintenance system according to claim 1, wherein, A safety valve is further installed on the steam curing chamber and connected to the control module, and the indoor environment acquisition module is further configured to acquire the pressure inside the steam curing chamber. When the pressure inside the steam curing chamber is greater than a preset threshold, the control module controls the safety valve to release the pressure of the steam curing chamber.

4. The precast T-beam intelligent temperature control maintenance system according to claim 1, wherein, The control module is provided with a temperature prediction unit and a temperature control unit, the temperature prediction unit is configured to predict the temperature inside the prefabricated T-beam at the next time according to the parameters acquired by the outdoor environment acquisition module, the indoor environment acquisition module, the steam generator parameter acquisition module, the spraying system parameter acquisition module, and the prefabricated T-beam temperature acquisition module, and the temperature control unit is configured to generate the opening degree parameters of the cooling water control valve and the steam control valve at the next time according to the temperature inside the prefabricated T-beam at the next time, and to control the steam flow of the steam generator and the cooling water flow of the spraying system.

5. The precast T-beam intelligent temperature control maintenance system according to claim 4, wherein, The temperature prediction unit comprises an LSTM prediction network, which comprises an input layer, two hidden layers, and an output layer, wherein the input feature vector of the input layer is a data sequence acquired at time t-1, including the temperature and wind speed of the environment outside the steam curing chamber at time t-1, the temperature and humidity of the environment inside the steam curing chamber at time t-1, the steam flow of the steam generator at time t-1, the cooling water flow of the spraying system at time t-1, and the temperature inside the prefabricated T-beam at time t-1; The hidden layer obtains the hidden state at time t according to the input feature vector of the input layer at time t-1. The output layer obtains the temperature inside the prefabricated T-beam at the time t according to the hidden state at the time t.

6. The precast T-beam intelligent temperature control maintenance system according to claim 5, wherein, The system introduces an attention mechanism in the LSTM prediction network, calculates an attention context vector according to the attention mechanism, and fuses the attention context vector and the hidden state at the time t through vector splicing and a full connection layer to form a state feature at the time t.

7. The precast T-beam intelligent temperature control maintenance system according to claim 6, wherein, The specific steps of introducing the attention mechanism in the LSTM prediction network to form the state feature at the time t are as follows: (1) Calculate the attention score of historical time i to current time t As follows: ； where, is the attention vector, T is the transpose, is the hidden state at time i is the weight of the hidden state, is the hidden state at time t is the weight of the hidden state, is the bias of the attention, tanh function outputs [-1, 1]; (2) the calculated attention score of the historical time i to the current time t is normalized, and the normalized attention score is as follows: ； wherein , and ; (3) Based on the normalized attention scores Computing attention context vectors as follows: ； (4) The LSTM prediction network outputs the hidden state at time t with the attention context vector Through vector splicing and full connection layer fusion, the state feature at time t is formed Specifically as follows: ； wherein, is a target deviation feature vector at time t, , is a fusion weight and bias, includes a temperature deviation at time t, a temperature rate of change deviation at time t, and a humidity deviation at time t.

8. The precast T-beam intelligent temperature control maintenance system according to claim 6, wherein, The temperature control unit comprises a DQP decision network, the DQP decision network sets a state space, an action space and a reward function, calculates a state-action value Q, and obtains an optimal action space according to the calculated state-action value Q, the action space comprising a valve opening degree of a steam control valve and a valve opening degree of a cooling water control valve.

9. The precast T-beam intelligent temperature control maintenance system according to claim 8, wherein, The steps of obtaining the optimal action space through the DQP decision network are as follows: (1) Defining the state space of a DQP decision network is: ； wherein is a phase identifier, , respectively, represent the phases of static rest, temperature increase, constant temperature, temperature decrease. The action space of the DQP decision network is defined as: ； wherein, is a valve opening degree of the steam control valve at time t, is a valve opening degree of the cooling water control valve at time t; (2) The calculation of the state-action value Q is as follows: ； wherein, are parameters of the DQP decision network, FC is a fully connected layer, and Q value represents the future cumulative reward at state Take action . (3) The reward function is set as follows: ； , , are weights, and + + = 1, and , is a temperature deviation at time t, is a temperature rate of change deviation at time t, is a humidity deviation at time t; (4) Adopting time series difference error to optimize network parameters and select the current maximum action space as the optimal action space with probability 1-ε.

10. The precast T-beam intelligent temperature control maintenance system according to claim 1, wherein, The prefabricated T-beam temperature acquisition module is also used to acquire the temperature outside the prefabricated T-beam, and the control module is also used to control the temperature outside the prefabricated T-beam to be less than a preset temperature difference from the temperature inside the prefabricated T-beam.

Images