Intelligent control method based on PID-dmc and big data modeling

Abstract

This invention relates to an intelligent control method based on PID-DMC and big data modeling. It calculates the control error between the target crude oil temperature and the actual measured crude oil temperature, as well as the rate of change of crude oil temperature. It determines whether the control error and the rate of change of temperature meet the requirements. If so, a recommended control quantity model is established to calculate the recommended control value of the gas flow rate, which is then input into the PID controller to control the crude oil temperature. Otherwise, by analyzing historical data, a model conforming to the step response is found and input into the DMC controller to control the crude oil temperature. The method acquires the temperature curve, determines the temperature inflection point, and establishes the heating start time and temperature control strategy. Control ends when the actual measured crude oil temperature reaches and remains stably at the target crude oil temperature within the required time. This invention solves the problem of untimely temperature control during crude oil transmission, avoids the energy waste and operational risks associated with traditional manual adjustment of the output temperature, and effectively solves the technical problems existing in the temperature control of current heating furnaces.

Description

Technical Field

[0001] This invention relates to the field of process control technology, and in particular to an intelligent control method based on PID-DMC and big data modeling. Background Technology

[0002] During pipeline transportation, crude oil containing wax has a high viscosity at room temperature and needs to be heated to reduce its viscosity before it can flow. However, there are target temperatures for heating crude oil. If the target temperature is too high, it wastes energy and may even lead to coking and failure of the heating furnace in wells with poor fluid supply capacity; while if the target temperature is too low, it can easily clog the pipeline. Traditionally, the external transport temperature is manually adjusted to heat the crude oil, but this depends on the skill and experience of the operators.

[0003] In actual operation, oilfield sites face complex and ever-changing working conditions and external disturbances. Timely responses are required to various process switching, changes in target heating temperature, and sudden changes in the flow rate of the heating medium, in order to achieve the desired production control effect.

[0004] During the production process, workers have fixed working hours, and specific production processes need to be carried out within these hours. For example, the hot washing process requires maintaining the outlet temperature of the heating medium water at 70°C. If the heating medium water reaches 70°C before the workers' working hours, it will waste energy; if it does not reach 70°C within the working hours, it will affect production.

[0005] Due to the long transmission pipelines and large heating furnace volume of oilfield stations, as well as the need to switch between different production processes, temperature control faces challenges including long lag times, severe interference, and the control of nonlinear objects. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by using a PID-DMC algorithm and big data modeling to solve the problem of untimely temperature control during crude oil transmission through pipelines. This avoids the energy waste and operational risks associated with traditional manual temperature adjustment. By combining big data modeling and consideration of multiple key points, the gas flow rate can be quickly adjusted to stabilize the temperature of the heating medium, water. By preheating the water and ensuring it reaches the target temperature at a specified time, the temperature can be maintained within a fixed working time. This satisfies production needs while saving energy, effectively solving the technical problems of temperature control in existing heating furnaces.

[0007] To achieve the above objectives, embodiments of the present invention provide an intelligent control method based on PID-DMC and big data modeling, the method comprising:

[0008] (1) Calculate the control error between the target value of crude oil temperature and the actual measured value of crude oil temperature, as well as the rate of change of crude oil temperature;

[0009] (2) Determine whether the control error is greater than the first threshold and whether the temperature change rate is greater than the second threshold. If yes, execute step (3); if no, execute step (4).

[0010] (3) Establish a recommended control quantity model for gas flow rate, calculate the recommended control value of gas flow rate through the recommended control quantity model, and input the recommended control value into the PID controller to control the crude oil temperature;

[0011] (4) Analyze historical data to find a model that meets the step response, and input the model into the DMC controller to control the crude oil temperature;

[0012] (5) Obtain the temperature curve from the historical temperature data and determine the temperature inflection point based on the temperature curve; determine the heating start time and temperature control strategy based on the temperature inflection point;

[0013] (6) Determine whether the actual measured value of the crude oil temperature has reached and is stably maintained at the target value of the crude oil temperature within the required time. If yes, end the control; otherwise, continue to execute step (1).

[0014] Preferably, the step of establishing a recommended control quantity model for gas flow rate and calculating the recommended control value of gas flow rate using the recommended control quantity model specifically includes:

[0015] (21) Select key points that affect crude oil temperature;

[0016] (22) Set the time window and sliding step size;

[0017] (23) Calculate the standard deviation of each key point within the time window;

[0018] (24) Determine whether the operating data of the time window is stable by using a stability threshold;

[0019] (25) Extract the average value within the stable time window; the average value includes the first average value and the second average value;

[0020] (26) Construct a recommendation control quantity model, and use the first average value as the input dataset of the recommendation control quantity model, and use the second average value as the output dataset of the recommendation control quantity model;

[0021] (27) Train the recommended control quantity model;

[0022] (28) Calculate the recommended control value of the gas flow rate using the recommended control quantity model.

[0023] Preferably, the step of obtaining a temperature curve from historical temperature data and determining the temperature inflection point based on the temperature curve specifically includes:

[0024] (31) Preprocess the historical temperature data to remove outliers and interpolate missing values ​​to obtain temperature curves;

[0025] (32) Calculate the time in the temperature curve from the initial time when the temperature difference with the initial time is a local maximum or local minimum;

[0026] (33) Calculate the temperature difference between the current moment and the moment of the local maximum or local minimum in the temperature curve. If the temperature difference is greater than the temperature transition threshold, record the current moment in the temperature transition array.

[0027] (34) Return to the temperature transition array, which stores temperature transition points.

[0028] More preferably, the method further includes:

[0029] The end time of the stable phase is calculated based on the temperature transition array, and the temperature curve is divided into the heating phase, cooling phase and stable phase.

[0030] More preferably, determining the heating start time through the temperature inflection point specifically includes:

[0031] (51) Statistical analysis of the heating time distribution data of crude oil under multiple gas flow rates and medium water flow rates;

[0032] (52) Construct a heating capacity model based on the heating time distribution data and obtain the heating capacity curve; the heating capacity curve is the curve of the maximum achievable temperature changing with the heating time;

[0033] (53) Design the expected trajectory of the heating process, determine the heating start time, and achieve the goal of heating the crude oil to the target temperature value at a specific time.

[0034] Preferably, the temperature control strategy includes a rapid heating mode and an energy-saving heating mode.

[0035] The intelligent control method based on PID-DMC and big data modeling provided in this invention first calculates the control error between the target crude oil temperature and the actual measured crude oil temperature, as well as the rate of change of crude oil temperature. If the control error is greater than a first threshold and the rate of change of temperature is greater than a second threshold, a recommended control quantity model for gas flow is established, and the recommended control value for gas flow is calculated using this model and input into the PID controller to control the crude oil temperature. If the control error is not greater than the first threshold or the rate of change of temperature is not greater than the second threshold, a model conforming to a step response is found through analysis of historical data and input into the DMC controller to control the crude oil temperature. Temperature curves are obtained from historical temperature data, and temperature inflection points are determined based on these curves. The heating start time and temperature control strategy are then determined based on these inflection points. When the actual measured crude oil temperature reaches and remains stably at the target crude oil temperature within the required time, control ends; otherwise, the above steps are repeated. This invention, based on the PID-DMC algorithm and big data modeling, solves the problem of untimely temperature control of crude oil during pipeline transmission, avoiding the energy waste and operational risks associated with traditional manual adjustment of the output temperature. It meets production needs while saving energy, effectively solving the technical problems existing in the temperature control of current heating furnaces. Attached Figure Description

[0036] Figure 1 A flowchart illustrating the intelligent control method based on PID-DMC and big data modeling provided in an embodiment of the present invention;

[0037] Figure 2 Temperature curves provided for embodiments of the present invention;

[0038] Figure 3 The expected temperature trajectory diagram of the heating process under different modes provided in the embodiments of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0040] This invention provides an intelligent control method based on PID-DMC and big data modeling. Figure 1 This is a flowchart illustrating the intelligent control method based on PID-DMC and big data modeling provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the intelligent control method based on PID-DMC and big data modeling of the present invention mainly includes the following steps:

[0041] Step (1): Calculate the control error between the target value of crude oil temperature and the actual measured value of crude oil temperature, as well as the rate of change of crude oil temperature.

[0042] In oilfield operations, conditions are highly complex and variable, involving various process switching and external disturbances. These changes include shifts in target temperature and sudden changes in the flow rate of the heating medium. To ensure effective production control, timely responses to these changes are necessary.

[0043] When the target crude oil temperature changes or the flow rate of the heating medium (e.g., water flow) changes abruptly, it usually leads to excessive control error or an excessively rapid rate of temperature change. In such cases, a PID algorithm is used to control the crude oil temperature. Therefore, it is necessary to monitor and calculate in real time the control error between the target crude oil temperature and the actual measured crude oil temperature, as well as the rate of temperature change of the crude oil, and to make judgments on the control error and the rate of temperature change.

[0044] Step (2): Determine whether the control error is greater than the first threshold and whether the temperature change rate is greater than the second threshold. If yes, proceed to step (3); otherwise, proceed to step (4).

[0045] Specifically, the first threshold is a pre-set error threshold, and the second threshold is a pre-set temperature change rate threshold. The method for controlling the crude oil temperature is determined by comparing the control error with the first threshold and the temperature change rate with the second threshold. When the control error is greater than the first threshold and the temperature change rate is greater than the second threshold, step (3) is executed, i.e., the crude oil temperature is controlled using the PID algorithm. When the control error is not greater than the first threshold or the temperature change rate is not greater than the second threshold, step (4) is executed, i.e., the crude oil temperature is controlled using the DMC algorithm.

[0046] Step (3): Establish a recommended control quantity model for gas flow rate, calculate the recommended control value for gas flow rate using the recommended control quantity model, and input the recommended control value into the PID controller to control the crude oil temperature.

[0047] Specifically, by combining big data, a recommended control model for gas flow is established to determine the gas flow rate required to maintain the crude oil temperature at the target temperature under different operating conditions. When the target crude oil temperature changes or the heating medium water flow rate changes abruptly, the model can be used to calculate the recommended control value of the gas flow rate, and the recommended control value can be used to replace the integral term of the PID controller in a timely manner to control the crude oil temperature, thereby avoiding excessively long oscillation adjustments during the control process.

[0048] The process of constructing the recommended control quantity model for gas flow is as follows:

[0049] Step (21): Select the key points that affect the temperature of crude oil at the outlet.

[0050] Specifically, the key points are generally PLC (Programmable Logic Controller) points. Based on requirements and system characteristics, determine which PLC points have a significant impact on crude oil temperature. This may include PLC points such as the upstream station's output temperature, the heating medium water flow rate, and the downstream station's inlet temperature. Let the set of key PLC points be K = {k1, k2, L, k n}, where k i This represents the i-th critical PLC point.

[0051] Step (22): Set the time window and sliding step size.

[0052] Specifically, let the time window length be T and the sliding step size be S, ensuring that T > control system lag time.

[0053] Step (23): Calculate the standard deviation of each key point within the time window.

[0054] Specifically, for each PLC point k i Calculate the standard deviation σ(k) of its value within the time window. i ).

[0055] Step (24): Determine whether the operating data of the time window is stable by using the stability threshold.

[0056] Specifically, a stability threshold θ is set; if for all critical PLC points k i They all have σ(k) i If θ ≤ θ, then the operating data for that time window is considered stable.

[0057] Step (25): Extract the average value within the stable time window.

[0058] Specifically, the average values ​​include the first average and the second average. For the j-th (j = 1, 2, ..., N) time window where the operating data is stable, extract k for each key PLC point. i The average value within this window is used as the first average value and forms the input feature vector x. j , represented as x j =(x 1j ,x 2j ,L,x nj ), where x ij This represents the average value of the i-th key PLC point within the j-th time window. Simultaneously, the average value of the control quantity within this time window is extracted as the second average value to form the output dataset {y}. j}

[0059] Step (26): Construct the recommended control quantity model.

[0060] Specifically, this model is a neural network model, using the extracted first average value as the input dataset {x} for the neural network model. j |j=1,2,L,N}, using the extracted second average as the output dataset {y} j |j=1,2,L,N}, construct an appropriate neural network model f(x)=y.

[0061] Step (27): Train the recommended control quantity model.

[0062] Specifically, the recommended control variable model is trained, and its parameters are tuned. The recommended control variable model f is trained using the constructed dataset, and its parameters are adjusted as needed to achieve a better fit.

[0063] Step (28): Calculate the recommended control value of gas flow rate using the recommended control quantity model.

[0064] Specifically, using a pre-trained recommended control quantity model f, the input is the changed target heating temperature or the abrupt change in the heating medium water flow rate. The predicted value of the corresponding control quantity is obtained from the model, denoted as y. pre =f(x) new ), where x new This indicates the new stable value that needs to be maintained at the critical PLC point, y pre This refers to the recommended control value for gas flow rate.

[0065] Step (4): Analyze historical data to find a model that matches the step response, and input the model into the DMC controller to control the crude oil temperature.

[0066] Step (5): Obtain the temperature curve from the historical temperature data and determine the temperature inflection point based on the temperature curve; determine the heating start time and temperature control strategy based on the temperature inflection point.

[0067] Specifically, in actual production, it is necessary to ensure that workers perform specific production processes according to fixed working hours, and that the crude oil temperature reaches and remains stable at a preset target temperature. To meet production demands while saving energy, big data modeling can be used to design different heating modes. This allows the heating furnace to heat the medium water at appropriate times, ensuring it reaches the target temperature at a specified moment and remains stable.

[0068] By analyzing temperature curves from historical temperature data, the curves are divided into heating, cooling, and steady-state portions. The key is to identify the inflection points between the heating and cooling phases. These inflection points help determine the appropriate heating start time and temperature control strategy.

[0069] The temperature curve and its temperature inflection points are determined using the following method:

[0070] Step (31) Data preprocessing: Preprocess the historical temperature data y(t) to obtain the temperature curve, including removing outliers and interpolating missing values ​​to ensure the integrity and consistency of the data.

[0071] Step (32): Calculate the time in the temperature curve where the temperature difference from the initial time is a local maximum or local minimum.

[0072] Step (33): Calculate the temperature difference between the current moment and the moment of the local maximum or local minimum in the temperature curve. If the temperature difference is greater than the temperature transition threshold, record the current moment in the temperature transition array.

[0073] Specifically, starting from the current value of k, y0 = y(k), k is added to the temperature transition array Z. The maximum and minimum values ​​of the distance along the vertical axis are y0 and y0. m Assign an initial value y m =0, and let x =0. m =k, the distance d along the vertical axis of the point comparison is given an initial value of d=0.

[0074] The algorithm iterates backwards to find the value y(k+i) at times k+i, i = 1, 2, L, and calculates d = ||y(k+i) - y(Z[-1])||, where Z[-1] represents the last element in the temperature transition array Z, i.e., the x-coordinate of the last transition point in array Z. If d > y m y will be updated m :=d,x m :=k+i. If ||dy m ||>e, where e is the positive threshold for temperature inflection, i.e., the temperature inflection threshold. Then the x-coordinate of the local inflection point is x. m , will x m Add to the temperature transition array Z, and update y at the same time. m :=0, x m = 0.

[0075] Step (34): Return the temperature transition array.

[0076] Specifically, after searching all points, the x-coordinates of the local temperature inflection points are stored in the temperature inflection array Z and returned, at which point the program terminates. The temperature inflection array Z stores all the temperature inflection points.

[0077] In industrial production, a water-mixing process is typically performed before the hot washing process to heat the medium with a low gas flow rate for an extended period. During this stage, the temperature of the medium is relatively stable and low. To determine the end time of this relatively stable temperature stage, the end time of the stable stage can be calculated based on a temperature transition array using the following calculation method:

[0078] Determine the search range: Obtain the first two elements {x0, x1} in the temperature transition array Z obtained above. x0 and x1 represent the starting time point of the relatively stable temperature stage and the x-coordinate of the first temperature transition point, respectively. Find the ending time point x of the relatively stable temperature stage between these two x-coordinates. en d.

[0079] Loop through intervals: For every integer value X∈{x0+i|i=1,2,L;x0+i<x1}=A in [x0,x1], divide the interval [x0,x1] and [y(x0),y(x1)] into intervals [x0,X], [X,x1], [y(x0),y(X)], and [y(X),y(x1)], respectively.

[0080] Data normalization: According to the normalization formula, Where x min =x0,x max =x1; in The intervals [x0,X] and [X,x1] are transformed into normalized intervals respectively. The intervals [y(x0), y(X)] and [y(X), y(x1)] are transformed into normalized values ​​respectively.

[0081] Constructing vectors: The origin of the rectangular coordinate system is O, and the coordinates of the normalized interval boundary points are denoted as follows: Let vector vector

[0082] Using the Law of Cosines to find angles: As the x-axis changes, ∠P1P2P3 also changes accordingly.

[0083] Determine the optimal solution at the end of the relatively stable temperature phase.

[0084] Figure 2 Temperature curves provided for embodiments of the present invention, such as Figure 2 As shown, by using the above methods, the temperature inflection point and the end time of the relatively stable temperature stage in the temperature curve can be found, thus distinguishing the heating stage, cooling stage and stable stage in the temperature curve.

[0085] The process of determining when to begin heating a medium to reach and maintain a stable target temperature is as follows:

[0086] In the previous method, the temperature inflection point and the end point of the relatively stable temperature stage were determined based on the temperature curve, and the temperature rise and fall intervals of the curve were also defined. Therefore, determining the heating start time through the temperature inflection point specifically includes:

[0087] Step (51): Statistically analyze the heating time distribution data of crude oil under multiple gas flow rates and medium water flow rates.

[0088] Step (52): Construct a heating capacity model based on the heating time distribution data, establish the relationship between the achievable maximum temperature and the heating time distribution under different operating conditions, and obtain the heating capacity curve. The heating capacity curve is the curve of the achievable maximum temperature changing with the heating time.

[0089] Step (53): By reasonably designing the expected trajectory of the heating process, the heating start time is determined, so as to heat the crude oil to the target temperature value at a specific time.

[0090] Figure 3 The expected temperature trajectory diagrams of the heating process under different modes provided in the embodiments of the present invention are as follows: Figure 3 As shown, the heating capacity curve is the curve showing the change in the achievable maximum temperature over time. The temperature control strategy includes a rapid heating mode and an energy-saving heating mode. In the rapid heating mode, the crude oil temperature quickly approaches the target temperature and then gradually stabilizes. In the energy-saving heating mode, the crude oil temperature rises slowly at first, then gradually increases, and finally slows down, at which point the crude oil temperature approaches the target temperature. Regardless of the mode, for any given temperature, the slope of the designed heating curve must be less than the slope of the heating capacity curve at the corresponding point at that same temperature.

[0091] In rapid heating mode, the expected temperature trajectory of the heating process can be represented by an exponential function as y = a·(1-exp(-b·t))+c, where a, b, and c are the three parameters to be determined in the exponential function; in energy-saving heating mode, the expected temperature trajectory of the heating process can be represented by the sigmoid function in the S-shaped curve. The appropriate transformation is expressed as Where a, b, c, and d are the four parameters to be determined in the sigmoid function. The rationale for verifying the reasonableness of the parameter settings is that, for any temperature, the slope of the designed heating curve is less than the slope of the heating capacity curve at the corresponding point at the same temperature. Once the parameter values ​​are determined, the expected heating trajectory is determined under this heating mode. During the control process, we hope that the actual temperature curve coincides with the expected temperature trajectory. However, the control system has a significant time delay. Therefore, in actual control, a strategy to simplify the control process can be adopted, selecting the expected temperatures {y(T), y(2T), L, y(kT)} at fixed intervals T over time intervals {T, 2T, L, kT} as the control target temperatures for the time intervals {[0, T), [T, 2T), L, [(k-1)T, kT)}.

[0092] Step (6): Determine whether the actual measured value of crude oil temperature has reached and is stably maintained at the target value of crude oil temperature within the required time. If yes, end the control; otherwise, continue to execute step (1).

[0093] Specifically, when applying the temperature control method provided by this invention, the actual measured value of crude oil temperature is continuously monitored. When the actual measured value of crude oil temperature reaches and is stably maintained at the target value of crude oil temperature within the required time, the temperature control method ends. If it has not yet reached the target value, the process continues from step (1).

[0094] This invention provides a method for calculating the control error between the target crude oil temperature and the actual measured crude oil temperature, as well as the rate of change of crude oil temperature. If the control error exceeds a first threshold and the rate of change of temperature exceeds a second threshold, a recommended control quantity model for the gas flow rate is established. The recommended control value for the gas flow rate is then calculated using this model and input into the PID controller to control the crude oil temperature. If the control error is not greater than the first threshold or the rate of change of temperature is not greater than the second threshold, a model conforming to a step response is identified through analysis of historical data and input into the DMC controller to control the crude oil temperature. Temperature curves are obtained from historical temperature data, and temperature inflection points are determined based on these curves. The heating start time and temperature control strategy are then determined based on these inflection points. When the actual measured crude oil temperature reaches and remains stably at the target crude oil temperature within the required timeframe, control is terminated; otherwise, the above steps are repeated. This invention, based on the PID-DMC algorithm and big data modeling, solves the problem of untimely temperature control during crude oil pipeline transmission, avoiding the energy waste and operational risks associated with traditional manual adjustment of the output temperature. It meets production needs while saving energy, effectively solving the technical problems existing in current heating furnace temperature control.

[0095] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0096] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0097] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent control method based on PID-DMC and big data modeling, characterized in that, The method includes: (1) Calculate the control error between the target value of crude oil temperature and the actual measured value of crude oil temperature, as well as the rate of change of crude oil temperature; (2) Determine whether the control error is greater than the first threshold and whether the temperature change rate is greater than the second threshold. If yes, execute step (3); if no, execute step (4). (3) Establish a recommended control quantity model for gas flow rate, calculate the recommended control value of gas flow rate through the recommended control quantity model, and input the recommended control value into the PID controller to control the crude oil temperature; (4) Analyze historical data to find a model that meets the step response, and input the model into the DMC controller to control the crude oil temperature; (5) Obtain the temperature curve from the historical temperature data and determine the temperature inflection point based on the temperature curve; determine the heating start time and temperature control strategy based on the temperature inflection point; (6) Determine whether the actual measured value of the crude oil temperature has reached and is stably maintained at the target value of the crude oil temperature within the required time. If yes, end the control; otherwise, continue to execute step (1).

2. The intelligent control method based on PID-DMC and big data modeling according to claim 1, characterized in that, The step of establishing a recommended control quantity model for gas flow rate and calculating the recommended control value of gas flow rate using the recommended control quantity model specifically includes: (21) Select key points that affect crude oil temperature; (22) Set the time window and sliding step size; (23) Calculate the standard deviation of each key point within the time window; (24) Determine whether the operating data of the time window is stable by using a stability threshold; (25) Extract the average value within the stable time window; the average value includes the first average value and the second average value; (26) Construct a recommendation control quantity model, and use the first average value as the input dataset of the recommendation control quantity model, and use the second average value as the output dataset of the recommendation control quantity model; (27) Train the recommended control quantity model; (28) Calculate the recommended control value of the gas flow rate using the recommended control quantity model.

3. The intelligent control method based on PID-DMC and big data modeling according to claim 1, characterized in that, The step of obtaining a temperature curve from historical temperature data and determining the temperature inflection point based on the temperature curve specifically includes: (31) Preprocess the historical temperature data to remove outliers and interpolate missing values ​​to obtain temperature curves; (32) Calculate the time in the temperature curve from the initial time when the temperature difference with the initial time is a local maximum or local minimum; (33) Calculate the temperature difference between the current moment and the moment of the local maximum or local minimum in the temperature curve. If the temperature difference is greater than the temperature transition threshold, record the current moment in the temperature transition array. (34) Return to the temperature transition array, which stores temperature transition points.

4. The intelligent control method based on PID-DMC and big data modeling according to claim 3, characterized in that, The method further includes: The end time of the stable phase is calculated based on the temperature transition array, and the temperature curve is divided into the heating phase, cooling phase and stable phase.

5. The intelligent control method based on PID-DMC and big data modeling according to claim 4, characterized in that, The process of determining the heating start time through the temperature inflection point specifically includes: (51) Statistical analysis of the heating time distribution data of crude oil under multiple gas flow rates and medium water flow rates; (52) Construct a heating capacity model based on the heating time distribution data and obtain the heating capacity curve; the heating capacity curve is the curve of the maximum achievable temperature changing with the heating time; (53) Design the expected trajectory of the heating process, determine the heating start time, and achieve the goal of heating the crude oil to the target temperature value at a specific time.

6. The intelligent control method based on PID-DMC and big data modeling according to claim 1, characterized in that, The temperature control strategy includes a rapid heating mode and an energy-saving heating mode.

Images