Temperature control method and system for high and low temperature impact heat flow meter

Abstract

The invention relates to the field of control, in particular to a temperature control method and system for a high and low temperature impact heat flow meter. Model training; the trained prediction model is utilized to calculate probability values of all possible observation state sequences from the current moment to the target moment when the hidden state is the target temperature, and the observation state sequences with the probability values larger than a probability threshold value are screened out to serve as candidates of the optimal adjustment parameter set; calculating the adjustment difficulty of each observation state sequence in the optimal adjustment parameter set, and selecting the observation state sequence with the maximum ratio of the probability value to the adjustment difficulty as an optimal parameter sequence; and adjusting parameter setting of the heat flow meter from the current moment to the target moment according to the optimal parameter sequence. According to the method, a group of optimal parameter adjustment sets can be dynamically found out through the prediction model according to the current heat flow instrument temperature and the target heat flow instrument temperature, so that accurate temperature control is realized.

Description

Technical Field

[0001] The present invention relates to the field of control, and more specifically, to a temperature control method and system for a high and low temperature impact heat flow instrument. Background Art

[0002] High and low temperature shock heat flow instrument is an experimental device used to simulate the thermal stress and thermal stability of materials under rapid temperature changes. It is used in the fields of materials science, electronics industry, automobiles, etc. It is of great significance for studying the thermal stability, thermal conductivity and thermal stress performance of new materials, coatings or composite materials, and detecting and verifying the adaptability of electronic components, optical components, aerospace equipment, etc. to high and low temperature shock. At the same time, it can also evaluate the effectiveness of thermal management systems when facing rapid temperature changes.

[0003] Prior art, such as a patent application document with publication number CN118550338A, discloses a temperature control method, a temperature control device, a temperature control equipment and a storage medium. The temperature control method includes: obtaining temperature adjustment information of the temperature control equipment at each set temperature within a preset time period; performing statistical processing on the temperature adjustment information at each set temperature, calculating the temperature increase ratio and the temperature decrease ratio, and determining the target parameters at each set temperature based on these ratios; using the target parameters and the set temperature to construct a prediction model, using a fuzzy logic method to map the target parameters and the set temperature to fuzzy sets respectively, generating fuzzy rules and their support, performing temperature prediction through the fuzzy rules, and adjusting the current set temperature of the temperature control equipment to the target temperature range according to the results of the prediction model.

[0004] However, in actual applications, temperature may change rapidly due to various factors (such as environmental changes, equipment status, etc.). The above patent application documents lack real-time temperature monitoring and feedback, resulting in the inability to adjust the temperature of the temperature control device in a timely and accurate manner. Summary of the invention

[0005] In order to solve the technical problem that the above temperature control method has limitations, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a temperature control method of a high and low temperature shock heat flow instrument comprises: Collecting historical parameter sequences of the heat flow meter at multiple historical moments; the historical parameter sequences include intake temperature, intake pressure, intake flow rate and internal temperature of the heat flow meter; inputting each historical parameter sequence into a preset prediction model to obtain a trained prediction model; Obtain the target temperature reached inside the heat flow meter within the target time, and collect the internal temperature of the heat flow meter at the current moment, use the internal temperature of the heat flow meter at the current moment as the initial hidden state of the prediction model, and calculate the probability value of all observation state sequences in the prediction model from the current moment to the target moment when the hidden state is the target temperature; use the observation state sequence with a probability value greater than a preset probability threshold as the optimal adjustment parameter set; Calculating the adjustment difficulty of each observation state sequence in the optimal adjustment parameter set, and taking the observation state sequence corresponding to the maximum value of the ratio of the probability value corresponding to the observation state sequence to the adjustment difficulty as the optimal parameter sequence; Adjust the parameters of the heat flow instrument from the current moment to the target moment according to the optimal parameter sequence.

[0007] The present invention can more accurately predict the changing trend of the internal temperature of the heat flow meter by collecting historical parameter sequences and training the prediction model. In the adjustment process, the internal temperature at the current moment is used as the initial hidden state, and the probability value from the current moment to the target moment is calculated, so as to find the best adjustment parameter set. The adjustment difficulty of each observation state sequence in the best adjustment parameter set is further calculated, and the ratio of the probability value to the adjustment difficulty is used as an evaluation index. The observation state sequence corresponding to the maximum ratio is selected as the best parameter sequence. This can minimize the adjustment difficulty and improve the adjustment efficiency while ensuring the temperature adjustment effect.

[0008] Preferably, adjusting the parameters of the heat flow instrument from the current moment to the target moment according to the optimal parameter sequence includes: The optimal parameter sequence is analyzed, and the inlet temperature, inlet pressure and inlet flow rate of the heat flow instrument are gradually adjusted according to the parameter setting values ​​obtained from the analysis.

[0009] By gradually adjusting parameters such as the inlet temperature, pressure and flow rate, the temperature change inside the heat flow meter can be made smoother, avoiding excessive temperature fluctuations caused by sudden changes in parameters, thereby ensuring the continuity and stability of temperature control during the entire adjustment process, allowing the heat flow meter to smoothly transition from the current temperature to the target temperature.

[0010] Preferably, the prediction model is a hidden Markov model; after the hidden Markov model training is completed, a hidden state transition probability matrix, an observation probability matrix and a hidden state initial probability distribution are obtained.

[0011] Preferably, the probability value of the observed state sequence satisfies the relationship: ; In the formula, From the current time to the target time, The probability value of the observed state sequence when the hidden state is the target temperature, For the A sequence of observed states, For the target time The internal temperature of the heat flow meter, is the target temperature, , and are the hidden state transition probability matrix, observation probability matrix and hidden state initial probability distribution of the hidden Markov model respectively. Represents probability.

[0012] By calculating the probability value of the observed state sequence under the target temperature hidden state, it is possible to accurately predict or diagnose whether the heat flow instrument is likely to reach the target temperature in the future or currently.

[0013] Preferably, the calculation process of adjusting the difficulty includes: Calculate the information entropy of the parameters contained in each observation state sequence in the optimal adjustment parameter set, and use the information entropy as the complexity of the corresponding parameter; Calculate the average complexity of all parameters in the observation state sequence to obtain the adjustment difficulty of the observation state sequence.

[0014] By calculating the information entropy of the parameters in each observation state sequence, the complexity of each parameter can be accurately quantified. The larger the information entropy, the higher the uncertainty of the parameter, that is, the more difficult it is to predict or control its value, which reflects the higher complexity of the parameter; calculating the average value of the complexity of all parameters can obtain the overall adjustment difficulty of the observation state sequence. This indicator helps to fully understand the complexity of the system or model and provide a basis for subsequent optimization or adjustment.

[0015] Preferably, the calculation process of adjusting the difficulty further includes: Calculate the information entropy of the parameters contained in each observation state sequence in the optimal adjustment parameter set; calculate the variance of the parameters contained in each observation state sequence in the optimal adjustment parameter set; The product of the information entropy and the variance is taken as the complexity of the parameter, and the average value of the complexity of all parameters in the observation state sequence is calculated to obtain the adjustment difficulty of the observation state sequence.

[0016] Preferably, the calculation process of adjusting the difficulty further includes: Calculate the synergy between the two parameters in the observed state sequence, and take the average value of all synergies in the observed state sequence as the synergy factor; The adjustment difficulty is obtained by normalizing the ratio of the average value of the complexity of all parameters in the observed state sequence to the synergy factor.

[0017] By calculating the synergy between two parameters in the observed state sequence, the degree of interaction between parameters can be quantified. High synergy indicates that there is a strong mutual influence between parameters, while low synergy indicates that the parameters are relatively independent. The adjustment difficulty index obtained by normalizing the ratio of the average complexity of the parameters to the synergy factor comprehensively considers the complexity of the parameters and the synergy between the parameters. This index can more comprehensively reflect the adjustment difficulty of the system and provide strong support for optimization decisions.

[0018] In a second aspect, a temperature control system of a high and low temperature shock heat flux instrument comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the temperature control method of the high and low temperature shock heat flux instrument is implemented.

[0019] The beneficial effects of the present invention are: The temperature control method of the present invention can dynamically find a set of optimal heat flow meter parameter adjustment sets according to the current heat flow meter temperature and the target heat flow meter temperature through a hidden Markov model, thereby achieving precise temperature control. This method not only takes into account the probability of reaching the target temperature, but also takes into account the complexity and synergy of the parameter sequence, ensuring the feasibility and effectiveness of parameter adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 It is a method flow chart of steps S1 to S5 in a temperature control method of a high and low temperature shock heat flow instrument according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] The embodiment of the present invention discloses a temperature control method of a high and low temperature impact heat flow instrument, referring to Figure 1 , including steps S1 to S5, which are specifically as follows: S1: Collecting a historical parameter sequence of a heat flow meter at multiple historical moments; the historical parameter sequence includes intake air temperature, intake air pressure, intake air flow rate, and internal temperature of the heat flow meter.

[0023] A heat flow meter is a device used to measure the heat flux density on the surface of an object. Its working principle is based on the basic physical laws of heat conduction and heat convection. In an impact heat flow meter, gas (usually air) is introduced into the instrument and impacts the sample under test through a specific channel, thereby generating heat flow. In this process, parameters such as gas temperature, pressure and flow rate will affect the temperature inside the heat flow meter.

[0024] The intake air temperature directly affects the temperature distribution inside the heat flow meter. When the intake air temperature rises, the temperature inside the heat flow meter will also rise accordingly. The influence of the intake air pressure on the temperature of the heat flow meter is mainly reflected in the gas flow state. A higher intake air pressure usually means that the gas flow is more intense, which will increase the heat exchange efficiency between the gas and the sample being tested. The intake air flow rate determines the amount of gas passing through the heat flow meter per unit time. A larger intake air flow rate means that more gas carries more heat energy into the heat flow meter.

[0025] In summary, the inlet temperature, inlet pressure and inlet flow are the key factors affecting the internal temperature of the heat flow meter. The changes in these parameters will directly affect the heat exchange efficiency between the gas and the sample being tested, thus causing the changes in the internal temperature of the heat flow meter. Therefore, by analyzing these inlet parameters, the temperature state of the heat flow meter can be indirectly understood.

[0026] Specifically, sensors are used to collect the intake temperature, intake pressure, intake flow rate and the temperature inside the heat flow meter at each moment (e.g., once per second) during the historical operation of the high and low temperature shock heat flow meter. That is, one timestamp corresponds to a parameter sequence, which includes the intake temperature, intake pressure, intake flow rate and internal temperature of the heat flow meter.

[0027] Furthermore, each of the above historical parameter sequences is normalized.

[0028] S2: Input each historical parameter sequence into the preset prediction model to obtain a trained prediction model.

[0029] In the embodiment of the present invention, the Hidden Markov Model (HMM) is selected as the prediction model of the temperature control method of the high and low temperature shock heat flow instrument, mainly based on the following reasons: 1. Separation of hidden state and observed state.

[0030] In the temperature control of the high and low temperature impact heat flow meter, the hidden state can be understood as the real temperature state inside the heat flow meter, which cannot be directly observed, but can be inferred through other parameters. Parameters such as intake temperature, intake pressure and intake flow rate can be directly observed, and they have a specific correlation with the temperature state of the heat flow meter.

[0031] 2. Modeling capabilities of sequence models.

[0032] Hidden Markov model is a statistical model of sequence data, which describes the transition probability between states and the emission probability of states to observations. In the temperature control of high and low temperature shock heat flow instrument, the temperature state and other parameters are sequence data that change over time. Hidden Markov model can capture the time dependence and correlation in these sequence data, so as to more accurately predict the future temperature state.

[0033] Specifically, the intake temperature, intake pressure and intake flow rate are used as the observed state of the hidden Markov model, and the internal temperature of the heat flow meter is used as the hidden state of the hidden Markov model. By training the hidden Markov model, the transition probability matrix of the hidden state, the probability matrix of the observed state and the initial probability distribution of the hidden state are obtained.

[0034] The intake temperature, intake pressure and intake flow rate at each moment are combined as an observation state sequence.

[0035] It should be noted that the hidden Markov model is an existing technology, and the specific training process will not be described in detail.

[0036] S3: Obtain the target temperature reached inside the heat flow meter within the target time, and collect the internal temperature of the heat flow meter at the current moment, use the internal temperature of the heat flow meter at the current moment as the initial hidden state of the prediction model, and calculate the probability value of all observation state sequences in the prediction model from the current moment to the target moment when the hidden state is the target temperature; use the observation state sequence with a probability value greater than a preset probability threshold as the optimal adjustment parameter set.

[0037] The temperature regulation of the heat flow meter is a complex dynamic process, which is affected by many factors, such as ambient temperature, heat conduction characteristics inside the heat flow meter, etc. The relationship between these factors is often nonlinear and uncertain, and it is difficult to directly describe it with a simple mathematical model. Hidden Markov chains can handle this uncertainty and complexity well, and capture the dynamic changes inside the system by introducing hidden states.

[0038] By using a temperature sensor to obtain the current temperature of the heat flow meter as the initial hidden state of the hidden Markov chain, the current state can be combined with historical data, and the state transition probability matrix and observation probability matrix in the hidden Markov model can be used to predict the possible path to reach the target temperature at the target time. This can more accurately reflect the state changes of the system at different time points and provide a basis for the selection of adjustment parameters.

[0039] Among them, the forward algorithm is an effective method for calculating the probability of occurrence of a given observation sequence in the hidden Markov model. Through the forward algorithm, the probability value of each observation state sequence from the current moment to the target moment can be calculated, thereby evaluating the possibility of reaching the target temperature under different adjustment parameter combinations.

[0040] Specifically, the target temperature reached inside the heat flow meter within the target time is obtained (i.e., the temperature value that the heat flow meter expects the inside of the heat flow meter to reach during the test, which is usually set by the staff according to specific test requirements or standards).

[0041] The internal temperature of the heat flow meter at the current moment is collected by the sensor, and the internal temperature of the heat flow meter at the current moment is used as the initial hidden state of the hidden Markov model, which means starting from this temperature, the dynamic change process of the internal temperature of the heat flow meter is simulated. In the process of model prediction, all possible observation state sequences or paths that allow the hidden state (i.e., the internal temperature of the heat flow meter) to reach the target temperature at the target moment are found.

[0042] Among them, the observed state is the state that can be directly observed in the hidden Markov model. For the heat flow meter, the observed state may include the intake temperature, intake pressure and intake flow measured by the temperature sensor. From the current moment to the target moment, there will be multiple possible observation state sequences, that is, different temperature change paths. Through the hidden Markov model, the probability value of each observation state sequence can be calculated, so as to evaluate the possibility of the heat flow meter reaching the target temperature from the current temperature under different conditions.

[0043] Then, by applying the forward algorithm, the probability values ​​of all observed state sequences in the hidden Markov model from the current moment to the target moment when the target hidden state is the target temperature are calculated, that is, the relationship is satisfied:

[0044] In the formula, From the current time to the target time, The probability value of the observed state sequence when the hidden state is the target temperature, For the A sequence of observed states, For the target time The internal temperature of the heat flow meter, is the target temperature, , and are the hidden state transition probability matrix, observation probability matrix and hidden state initial probability distribution of the hidden Markov model respectively. Represents probability.

[0045] in, Indicates that at the target time , the temperature inside the heat flow instrument reached the target temperature.

[0046] According to the above The calculation method of is similar to the calculation method to obtain the probability value of all observed state sequences from the current moment to the target moment when the hidden state is the target temperature.

[0047] The probability values ​​calculated by the forward algorithm provide quantitative information about the relationship between the observation sequence and the hidden state. These probability values ​​can be interpreted as the confidence that the system is in a specific hidden state (such as the target temperature) under a given observation sequence. Furthermore, setting a probability threshold can help screen out those observation state sequences that are most likely to lead to the target hidden state (target temperature). These sequences can be regarded as the best set of adjustment parameters because they are most likely to achieve the desired system state.

[0048] Specifically, the probability threshold is set to 0.6 (in practical applications, the choice of the threshold can be adjusted according to specific needs. For example, if the system has high requirements for accuracy, a higher threshold can be set to ensure that only the most likely observation state sequence is selected; on the contrary, if the system requires more flexibility or robustness, a lower threshold can be set to include more candidate sequences), and the observation state sequence with a probability value greater than the probability threshold is used as the optimal adjustment parameter set.

[0049] In general, traditional temperature control methods may require repeated trials and adjustments to find the right parameters, which is time-consuming and inefficient. However, using hidden Markov chains and forward algorithms, multiple possible combinations of adjustment parameters and their probability values ​​can be calculated in a short period of time, and the best solution can be quickly screened out, improving the efficiency and accuracy of the adjustment.

[0050] S4: Calculate the adjustment difficulty of each observation state sequence in the optimal adjustment parameter set, and take the maximum value of the ratio of the probability value corresponding to the observation state sequence to the adjustment difficulty as the optimal parameter sequence.

[0051] In each observed state sequence of the optimal adjustment parameter set, different parameters may have different impacts on the system and adjustment complexity. Therefore, by calculating the complexity of each parameter, we can better understand the importance of each parameter.

[0052] First, select any observation state sequence, and calculate the complexity of the parameters in the observation state sequence by analyzing the information entropy. The complexity of the parameters satisfies the relationship:

[0053] In the formula, For the The first observation state sequence The complexity of the parameters, For the The first observation state sequence The probability of the value corresponding to the parameter appearing is is the total number of observed state sequences in the optimal adjustment parameter set, is a logarithmic function (exemplarily, the base may be 10).

[0054] Then, according to the above calculation The first observation state sequence The complexity of each parameter can be used to calculate the complexity of other parameters in the observation state sequence. After obtaining the complexity of each parameter, these complexities are averaged to obtain the adjustment difficulty of the observation state sequence.

[0055] In another embodiment, the variance of the parameter value is introduced to calculate the complexity of the parameter, and the average value of the complexity of all parameters is used as the adjustment difficulty of the observed state sequence.

[0056] Among them, the complexity of the parameters after introducing the variance of the parameter value satisfies the relationship:

[0057] In the formula, For the The first observation state sequence The complexity of the parameters, For the The first observation state sequence The probability of the value corresponding to the parameter appearing is is the total number of observed state sequences in the optimal adjustment parameter set, For the The variance corresponding to the parameter.

[0058] It should be noted that different parameters often influence each other. Therefore, analyzing the synergy between parameters helps to develop more effective parameter adjustment strategies.

[0059] First, the synergy between the two parameters in the observed state sequence is calculated.

[0060] Exemplarily, the two parameters of intake temperature and intake pressure are selected, and the synergy between the two is calculated, that is, the relationship is satisfied:

[0061] In the formula, Intake air temperature With intake pressure The synergy between Intake air temperature With intake pressure The covariance between Intake air temperature The standard deviation of Intake pressure The standard deviation of .

[0062] When the synergy between parameters is high, it means that the adjustment trends of these parameters are relatively consistent, and the overall stability of the system may be higher. On the contrary, if the synergy between parameters is low, the stability of the system may be poor, because the adjustment of parameters may cause drastic changes in the system state.

[0063] According to the calculation method of the synergy between the intake temperature and the intake pressure, all the synergies in the observed state sequence can be obtained in the same way, and the average value of all the synergies is further calculated to obtain the synergy factor.

[0064] Furthermore, by combining the complexity of a single parameter and the synergy between multiple parameters, the overall adjustment difficulty of the observation state sequence is calculated, that is, the relationship is satisfied:

[0065] In the formula, is the adjustment difficulty of the observed state sequence, is the average value of the complexity corresponding to all parameters in the observed state sequence, is the average value of the synergy between parameters in the observed state sequence, For normalization processing.

[0066] In the above In the calculation formula, the difficulty of adjustment is closely related to the discreteness of the parameter values ​​and the interaction between the parameters. The larger the variance, the more discrete the distribution of the parameter values ​​and the greater the difference between the parameters. If the parameters are independent of each other, then each parameter can be considered separately when adjusting, which is relatively simple. However, if there is an interaction or dependency between the parameters, adjusting one parameter may affect the effects of other parameters, so it is necessary to consider the coordination and balance of multiple parameters at the same time.

[0067] According to the calculation method of the adjustment difficulty of the above observation state sequence, the adjustment difficulty of other observation state sequences in the optimal adjustment parameter set can be calculated in the same way.

[0068] In summary, the probability value corresponding to each observation state sequence in the optimal adjustment parameter set reflects the possibility of the corresponding sequence appearing in the prediction model. The higher the probability value, the more likely the sequence is to occur under the given model, which means that the model fits the sequence better. The difficulty of adjusting the observation state sequence reflects the complexity required to adjust the parameters to a specific observation state sequence.

[0069] Comprehensive consideration of probability and adjustment difficulty can make the model more operable and practical while maintaining good performance.

[0070] Specifically, the observed state sequence corresponding to the maximum value of the ratio of the probability value corresponding to the observed state sequence to the adjustment difficulty is taken as the optimal parameter sequence. The optimal parameter sequence represents the optimal adjustment path of each parameter (intake temperature, intake pressure, intake flow) of the heat flow instrument from the current moment to the target moment.

[0071] By calculating the ratio of probability value to adjustment difficulty, these two factors can be balanced to a certain extent. Selecting the observed state sequence with the largest ratio means minimizing the adjustment difficulty while ensuring a high probability, thus achieving a balance between the model fitting effect and the feasibility of parameter adjustment.

[0072] S5: Adjust the parameters of the heat flow instrument from the current moment to the target moment according to the optimal parameter sequence.

[0073] The optimal parameter sequence is analyzed to obtain the specific setting values ​​of each parameter of the heat flow instrument given in the optimal parameter sequence. These setting values ​​will guide the parameter adjustment of the heat flow instrument in actual operation.

[0074] Starting from the current moment, gradually adjust the parameters of the heat flow instrument. During and after the adjustment, continue to monitor the operating status of the heat flow instrument, especially the changes in internal temperature. If it is found that the actual parameters deviate significantly from the target parameters, or the performance of the heat flow instrument does not meet expectations, additional adjustments or diagnosis may be required.

[0075] An embodiment of the present invention also discloses a temperature control system for a high and low temperature shock heat flux instrument, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a temperature control method for the high and low temperature shock heat flux instrument according to the present invention is implemented.

[0076] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0077] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0078] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0079] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A temperature control method for a high and low temperature impact heat flow instrument, characterized in that: include: Collecting historical parameter sequences of the heat flow meter at multiple historical moments; the historical parameter sequences include intake temperature, intake pressure, intake flow rate and internal temperature of the heat flow meter; inputting each historical parameter sequence into a preset prediction model to obtain a trained prediction model; Obtain the target temperature reached inside the heat flow meter within the target time, and collect the internal temperature of the heat flow meter at the current moment, use the internal temperature of the heat flow meter at the current moment as the initial hidden state of the prediction model, and calculate the probability value of all observation state sequences in the prediction model from the current moment to the target moment when the hidden state is the target temperature; use the observation state sequence with a probability value greater than a preset probability threshold as the optimal adjustment parameter set; Calculating the adjustment difficulty of each observation state sequence in the optimal adjustment parameter set, and taking the observation state sequence corresponding to the maximum value of the ratio of the probability value corresponding to the observation state sequence to the adjustment difficulty as the optimal parameter sequence; Adjust the parameters of the heat flow instrument from the current moment to the target moment according to the optimal parameter sequence.

2. The temperature control method of a high and low temperature impact heat flow instrument according to claim 1, characterized in that: The step of adjusting the parameters of the heat flow instrument from the current moment to the target moment according to the optimal parameter sequence includes: The optimal parameter sequence is analyzed, and the inlet temperature, inlet pressure and inlet flow rate of the heat flow instrument are gradually adjusted according to the parameter setting values ​​obtained from the analysis.

3. The temperature control method of a high and low temperature impact heat flow instrument according to claim 2, characterized in that: The prediction model is a hidden Markov model; after the hidden Markov model training is completed, a hidden state transition probability matrix, an observation probability matrix and a hidden state initial probability distribution are obtained.

4. The temperature control method of a high and low temperature impact heat flow instrument according to claim 3, characterized in that: The probability value of the observed state sequence satisfies the relationship: ; In the formula, From the current time to the target time, The probability value of the observed state sequence when the hidden state is the target temperature, For the A sequence of observed states, For the target time The internal temperature of the heat flow meter, is the target temperature, , and are the hidden state transition probability matrix, observation probability matrix and hidden state initial probability distribution of the hidden Markov model respectively. Represents probability.

5. The temperature control method of a high and low temperature impact heat flow instrument according to claim 4, characterized in that: The calculation process of adjusting the difficulty includes: Calculate the information entropy of the parameters contained in each observation state sequence in the optimal adjustment parameter set, and use the information entropy as the complexity of the corresponding parameter; Calculate the average complexity of all parameters in the observation state sequence to obtain the adjustment difficulty of the observation state sequence.

6. The temperature control method of a high and low temperature impact heat flow instrument according to claim 5, characterized in that: The calculation process of adjusting the difficulty also includes: Calculate the information entropy of the parameters contained in each observation state sequence in the optimal adjustment parameter set; calculate the variance of the parameters contained in each observation state sequence in the optimal adjustment parameter set; The product of the information entropy and the variance is taken as the complexity of the parameter, and the average value of the complexity of all parameters in the observation state sequence is calculated to obtain the adjustment difficulty of the observation state sequence.

7. The temperature control method of a high and low temperature impact heat flow instrument according to claim 6, characterized in that: The calculation process of adjusting the difficulty also includes: Calculate the synergy between the two parameters in the observed state sequence, and take the average value of all synergies in the observed state sequence as the synergy factor; The adjustment difficulty is obtained by normalizing the ratio of the average value of the complexity of all parameters in the observed state sequence to the synergy factor.

8. A temperature control system for a high and low temperature impact heat flow instrument, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the temperature control method of the high and low temperature shock heat flow instrument according to any one of claims 1 to 7 is implemented.

Images