Energy-saving control method and system for medium-frequency heating based on data analysis

Abstract

The application relates to the technical field of heating control, in particular to a medium-frequency heating energy-saving control method and system based on data analysis, which specifically comprises the following steps: filtering the historical records of a heating process of an oil pumping unit in a medium-frequency heating process according to temperature changes of a heating area, consumed electric energy and heating duration; constructing a load-power fitting curve by using the load and equipment power data in the filtered historical records, and obtaining the theoretical optimal power under each load; comparing the working condition at the starting moment of a new heating cycle with the historical records, and setting the starting power; calculating the temperature rise efficiency and dynamically adjusting the heating power based on the difference between the actual power and the theoretical optimal power, and then performing medium-frequency heating; the temperature overshoot problem commonly existing in traditional control is avoided, the energy loss caused by overheating is reduced, and the safety of the heating process and the stability of crude oil treatment are improved.

Description

Technical Field

[0001] This application relates to the field of heating control technology, specifically to a data analysis-based medium-frequency heating energy-saving control method and system. Background Technology

[0002] In the oil extraction field, especially for crude oil with high wax content and high viscosity, wax deposition is prone to occur in the tubing and wellbore during pumping unit operation. This increases the resistance of the sucker rod movement and, in severe cases, can even lead to pump jamming and production shutdowns. Medium-frequency heating technology utilizes electromagnetic induction to heat the tubing, effectively raising the crude oil temperature, preventing wax crystal formation, reducing crude oil viscosity, and ensuring the normal operation of the pumping unit. Currently, medium-frequency heating equipment is widely used in oilfields. However, current medium-frequency heating control methods have technical shortcomings: during the continuous heating process of a single pumping stroke, the optimal heating power required changes in real time as the crude oil temperature rises and viscosity decreases. Existing control methods are mostly constant power heating or simple two-stage control, which cannot adjust the power in real time based on feedback information such as load and temperature rise efficiency, easily leading to insufficient or excessive heating.

[0003] Therefore, developing a medium-frequency heating energy-saving method and system that can perform precise and adaptive control based on historical data and real-time operating conditions is of great significance for improving the energy efficiency and automation level of oilfield production. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a data analysis-based energy-saving control method and system for medium-frequency heating, the specific technical solution of which is as follows: In a first aspect, embodiments of this application provide a medium-frequency heating energy-saving control method based on data analysis, the method comprising the following steps: Step S1: Take each continuous heating process in the medium-frequency heating of the oil pumping unit as a heating cycle; take the data within a heating cycle as a historical record; Step S2: Based on the equivalent temperature change of the medium-frequency heating area, the power consumption of heating, and the heating time in each heating cycle, filter each valid historical record and the excellent historical records among the valid historical records; Step S3: Construct a fitting curve of load versus power using the pumping unit suspension point load data and medium-frequency heating equipment power data from excellent historical records, and obtain the theoretical optimal power under each load based on the curve; Step S4: At the start of a new heating cycle, compare the operating condition data at the start of the new heating cycle with the operating condition data in each valid historical record, and set the starting power of the new heating cycle. Step S5: Calculate the temperature rise efficiency at each moment based on the equivalent temperature change of the heating area and the power consumption, and dynamically adjust the power of the new heating cycle by combining the actual power and the theoretical optimal power obtained based on the actual load. Step S6: Perform medium-frequency heating based on the dynamically adjusted power.

[0005] Furthermore, in step S2, the filtering of each valid historical record and the excellent historical records among the valid historical records includes: S21: If the equivalent temperature of the heating area is greater than the preset target temperature at the end of each heating cycle, then the historical records of each heating cycle shall be regarded as valid historical records. S22: Calculate the unit energy consumption temperature rise for each heating cycle based on the equivalent temperature rise and energy consumption of the heating area in each heating cycle; set the theoretical optimal heating time and compare it with the heating time of each heating cycle to calculate the heating time matching degree of each heating cycle; determine the temperature overshoot of each heating cycle based on the difference between the maximum equivalent temperature and the target temperature in each heating cycle; calculate the comprehensive score of each effective historical record based on the unit energy consumption temperature rise, the heating time matching degree, and the temperature overshoot. S23: Select the preset percentage of historical records with the highest overall scores from all valid historical records as excellent historical records.

[0006] Furthermore, the theoretically optimal heating time mentioned in step S22 is specifically set as the heating time of the heating cycle in which the historically equivalent temperature reaches the target temperature and the energy consumption is the lowest. The specific process for calculating the heating time matching degree is as follows: calculate the absolute value of the difference between the heating duration of each heating cycle and the theoretical optimal heating duration, calculate the ratio of the absolute value of the difference to the theoretical optimal heating duration, and take the difference between the constant 1 and the ratio as the heating time matching degree; Specifically, the temperature overshoot is defined as follows: if the difference between the maximum equivalent temperature and the target temperature is greater than 0, then the difference is taken as the temperature overshoot; otherwise, the temperature overshoot is set to 0.

[0007] Furthermore, the comprehensive score is a weighted sum of the negative correlation mapping value of the temperature overshoot, the temperature rise per unit energy consumption, and the matching degree of the heating time.

[0008] Furthermore, setting the starting power for the new heating cycle in step S4 includes: S41: The operating data includes ambient temperature, atmospheric humidity, and suspension point load; The vector composed of the absolute values ​​of ambient temperature and atmospheric humidity is used as the environmental parameter vector; the similarity between the environmental parameter vector at the start of the new heating cycle and the environmental parameter vector at the start of each valid historical record is calculated. S42: The difference between the load at the start of the new heating cycle and the load at the start of each valid historical record is processed to eliminate dimensions, and the load difference is obtained. S43: Based on the comprehensive score of each valid historical record, the similarity, and the load difference, determine the weighting factor for each valid historical record; S44: Calculate the product of the intermediate frequency heating power at the start time of each valid historical record and the weighting factor, and calculate the ratio of the sum of the products of all valid historical records to the sum of the weighting factors, as the starting power of the new heating cycle.

[0009] Furthermore, the calculation process of the weighting factor in step S43 is as follows: The product of the similarity and the comprehensive score of each valid historical record is used as the numerator, the sum of the load difference and the preset minimum positive number is used as the denominator, and the ratio of the numerator to the denominator is used as the weighting factor.

[0010] Furthermore, dynamically adjusting the power of the new heating cycle in step S5 includes: S51: When heating to any moment in a new heating cycle, take the average load within 1 second before the moment as the representative load of the moment, input the fitting curve to obtain the corresponding theoretical optimal power, and record it as the representative power; take the ratio of the power of the moment before the moment to the representative power as the load-power fit degree of the moment. S52: Divide the difference between the equivalent temperature of the heating area at each moment in each heating cycle and the equivalent temperature of the heating area at the beginning of the heating cycle by the total electrical energy consumed from the beginning to each moment of the heating cycle to obtain the real-time temperature rise efficiency at each moment in each heating cycle. S53: Obtain the temperature rise efficiency of each heating cycle, and based on the difference between the real-time temperature rise efficiency at any moment in the new heating cycle and the temperature rise efficiency of the heating cycle with the closest load, combine the load-power adaptability to obtain the dynamic power adjustment coefficient at any moment. S54: The product of the theoretical optimal power under load at any given moment and the dynamic power adjustment coefficient is taken as the dynamically adjusted power at any given moment in the new heating cycle.

[0011] Furthermore, the expression for the dynamic power regulation coefficient in step S53 is as follows: If the cumulative energy consumption from the start of the new heating cycle to any given time does not exceed the preset energy consumption threshold, then the dynamic power adjustment coefficient at time t in the new heating cycle is... for: ; otherwise, ; In the formula, The load-power fit at time t. Let be the real-time temperature rise efficiency at time t. The temperature rise efficiency is the heating cycle that is closest to the load at time t. , These are the preset first gain coefficient and second gain coefficient, respectively.

[0012] Furthermore, in step S6, performing medium-frequency heating based on the dynamically adjusted power includes: The power of the medium-frequency heating device is adjusted once every preset control cycle; when the equivalent temperature of the heating area exceeds the preset temperature threshold, the heating power is reduced to a preset multiple of the rated power of the medium-frequency heating device, wherein the preset multiple is greater than 0 and less than 1; when the equivalent temperature of the heating area exceeds the preset target temperature for a longer than a preset duration, heating is stopped.

[0013] Secondly, embodiments of this application also provide a medium-frequency heating energy-saving control system based on data analysis, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0014] Compared with the prior art, the technical solution adopted in this application has the following advantages: This application selects "excellent historical records" from massive amounts of historical operating data and establishes a fitting function between the suspension point load and the heating power. Based on this, it combines intelligent initial power estimation and dynamic power correction to ensure that the heating power always matches the actual load requirements throughout the process, effectively avoiding power redundancy and energy waste. Simultaneously, in the initial heating stage, it intelligently calculates a reasonable initial power using scores from historical successful cases, environmental similarity, and load differences, avoiding power surges caused by starting from zero or fixed power. Furthermore, a smooth dynamic adjustment coefficient is used during the heating process, making power adjustments gradual and continuous, reducing the electrical stress on the intermediate frequency power supply and heating equipment caused by power surges, and helping to extend the equipment's service life. In addition to dynamic power regulation, a temperature safety constraint mechanism is introduced. When the real-time equivalent temperature approaches the target temperature, the heating equipment automatically switches to a low-power maintenance mode, avoiding the temperature overshoot problem common in traditional control, reducing energy loss due to overheating, and improving the safety of the heating process and the stability of crude oil processing. Attached Figure Description

[0015] Figure 1 A flowchart illustrating the steps of a data analysis-based medium-frequency heating energy-saving control method provided in one embodiment of this application. Detailed Implementation

[0016] The following description, in conjunction with the accompanying drawings, details the specific scheme of the data analysis-based medium-frequency heating energy-saving control method and system provided in this application.

[0017] Please see Figure 1 The diagram illustrates a flowchart of a data analysis-based medium-frequency heating energy-saving control method according to an embodiment of this application. The method includes the following steps: Step S1: Take each continuous heating process in the medium-frequency heating of the oil pumping unit as a heating cycle; take the data within a heating cycle as a historical record.

[0018] This application aims to address the medium-frequency heating scenario of oil pumping units by adaptively optimizing heating power through in-depth analysis of historical operating data and real-time operating condition data, thereby achieving energy saving, consumption reduction, and precise temperature control in the medium-frequency heating process of oil pumping units.

[0019] This application first obtains historical operating condition data of the pumping unit's medium-frequency heating from the storage unit. In this embodiment, the obtained historical operating condition data is the operating condition data of the most recent N heating cycles. The types of operating condition data include: pumping unit suspension point load (unit: kN), operating current (unit: A) and operating voltage (unit: V) of the medium-frequency heating equipment, equivalent temperature of the medium-frequency heating area (unit: °C), duration of each continuous heating by the medium-frequency heating equipment (unit: s), ambient temperature, and atmospheric humidity.

[0020] In this embodiment, N is set to 5. In other embodiments of this application, the implementer may set the value of N according to the actual situation. Each continuous heating process of the pumping unit's intermediate frequency heating equipment is considered as a heating cycle. Historical data is divided according to independent heating cycles, and the heating cycle is not bound to a single mechanical stroke of the pumping unit. The data collected in each heating cycle is the same historical record.

[0021] Before acquiring operating condition data, this application first deploys a sensor network by installing various sensors on key parts of the pumping unit's intermediate frequency heating equipment. Specifically, in this embodiment, a load sensor installed at the suspension point measures the load at the pumping unit's suspension point; a current / voltage sensor connected to the intermediate frequency heating power supply line monitors instantaneous current and voltage values; a heat balance mathematical model is constructed to calculate the equivalent temperature of the heating area based on the intermediate frequency heating input power and ambient temperature parameters. The heat balance data model is a well-known technology, and the specific process will not be elaborated further; a timing module integrated into the main control unit or an independent timer records the heating duration of the heating cycle; and outdoor temperature and humidity sensors are installed in the well site control cabinet or near the pumping unit base to acquire ambient temperature and atmospheric humidity data. The sensor network synchronously collects various operating condition data at a fixed sampling frequency; in this embodiment, the sampling frequency is 10 times per second.

[0022] The analog signals output by each sensor are then converted into digital quantities by an analog-to-digital converter after signal conditioning circuitry, and temporarily stored in the buffer of the sensor node, forming timestamped data frames. Data packets collected by each sensor are uploaded to the main control unit in real time using wired (e.g., RS-485, CAN bus) or wireless (e.g., Wi-Fi, ZigBee) communication methods. The transmission protocol has a verification mechanism (e.g., CRC check) to ensure data integrity and reliability. After receiving all data, the main control unit aligns and merges them according to the time axis to form a multidimensional dataset containing "time-load-current-voltage-equivalent temperature-heating time," which is then stored in the storage unit. Historical operating condition data of the pumping unit's intermediate frequency heating can then be retrieved from the storage unit.

[0023] Step S2: Based on the equivalent temperature change of the medium-frequency heating area, the power consumption of heating, and the heating time in each heating cycle, filter each valid historical record and the excellent historical records among the valid historical records.

[0024] In one embodiment of this application, step S2 specifically includes: S21: If the equivalent temperature of the heating area is greater than the preset target temperature at the end of each heating cycle, then the historical records of each heating cycle are considered valid historical records; otherwise, they are considered invalid historical records. In this embodiment, the target temperature is set as the sum of the wax precipitation point temperature and the tolerance temperature H. Preferably, H is set to 5 in this embodiment. The wax precipitation point is a petroleum industry term referring to the critical temperature at which wax crystals begin to precipitate in crude oil, which can be obtained through prior knowledge.

[0025] The core purpose of medium-frequency heating is to heat the crude oil in the tubing through electromagnetic induction, preventing wax formation and reducing viscosity, thereby ensuring the normal operation of the pumping unit. However, during the heating process, excessive power can easily cause temperature overshoot (the actual temperature significantly exceeds the target temperature), resulting in a large amount of wasted electrical energy; insufficient power will lead to low heating efficiency, causing wax formation in the tubing and increased crude oil viscosity, which in turn leads to increased suspension load on the pumping unit and increased operating resistance, and in severe cases may cause accidents such as pump jamming and production shutdown.

[0026] However, not all of the massive historical data obtained is of reference value. Therefore, this application selects valid historical records to obtain samples with high heating efficiency, low energy consumption, and satisfactory performance to guide subsequent power adjustments.

[0027] S22: Calculate the unit energy consumption temperature rise for each heating cycle based on the equivalent temperature rise and energy consumption of the heating area in each heating cycle; set the theoretical optimal heating time and compare it with the heating time of each heating cycle to calculate the heating time matching degree of each heating cycle; determine the temperature overshoot of each heating cycle based on the difference between the maximum equivalent temperature and the target temperature in each heating cycle; calculate the comprehensive score of each valid historical record based on the unit energy consumption temperature rise, the heating time matching degree, and the temperature overshoot.

[0028] As a preferred embodiment of this application, the formula for calculating the comprehensive score is as follows: In the formula, The overall score for the a-th valid historical record; It represents the equivalent temperature rise of the heating region during the a-th heating cycle, which is the absolute value of the equivalent temperature difference between the end and start times of the heating cycle. Let be the total electrical energy consumed by the pumping unit's medium-frequency heating equipment during the heating process in the a-th heating cycle, denoted as electrical energy consumption. The heating duration of the a-th heating cycle; To determine the theoretically optimal heating time, the heating cycle with the lowest equivalent temperature at the end of the heating cycle and the lowest power consumption is obtained, and its heating time is taken as the theoretically optimal heating time. It should be noted that if there are multiple heating cycles with the lowest power consumption, the average heating time is taken. The temperature overshoot for the a-th heating cycle is calculated by obtaining the difference between the maximum equivalent temperature and the target temperature during the heating cycle. If the difference is greater than 0, the difference is taken as the temperature overshoot; otherwise, the temperature overshoot is set to 0. H is the preset tolerance temperature. , and These are the first, second, and third weighting factors, which are set in this application. The implementer can adjust it as needed to meet the requirements. That's all; This is a function to find the maximum value. This is a normalization function. The electrical energy is obtained by integrating the operating current and operating voltage over time, which is a well-known technique and will not be elaborated further.

[0029] This represents the temperature rise per unit of energy consumption; the higher the value, the higher the efficiency of converting electrical energy into heat energy and being absorbed by crude oil. For The normalization in this embodiment of the application is achieved by normalizing the temperature rise per unit energy consumption obtained from all historical records using the maximum and minimum values. Normalization. It should be noted that there are many existing normalization methods, and implementers can also use other normalization functions. Normalization is not subject to specific restrictions in this application.

[0030] This indicates the heating time matching degree, which measures how close the actual heating time is to the "ideal duration". This item penalizes time deviations, prompting the selection of an appropriate heating cycle to avoid unnecessary heat loss due to excessive heating time or insufficient heating time resulting in failure to reach the target temperature.

[0031] A larger temperature overshoot indicates that the heating system failed to stop in time after reaching the target, continuing to input electrical energy and causing the temperature to exceed the limit. In other words, the greater the energy waste, the lower the overall score. Comparing this to H, we obtain the specific percentage of scores exceeding the limit. Further subtracting this percentage from 1 yields the linearly distributed decrease in scores. The outermost layer of the formula... The operator compares the internal calculation result with 0 once, extracts the larger value between the two, and ensures that the minimum score for this item is 0.

[0032] S23: A predetermined percentage of all valid historical records with the highest overall scores are designated as excellent historical records. Further, in this embodiment, the percentage is 20%, meaning the top 20% of valid historical records sorted by overall score from highest to lowest are considered excellent historical records. Implementers can set this percentage according to their actual circumstances; this application does not impose specific restrictions.

[0033] Step S3: Construct a fitting curve of load versus power using the pumping unit suspension point load data and medium-frequency heating equipment power data from excellent historical records, and obtain the theoretical optimal power under each load based on the curve.

[0034] As an optional embodiment of this application, the construction of the fitting curve regarding load and power includes: S31: The average value of all suspension point load data in each historical record is taken as the suspension point load characteristic value of each historical record, and the average value of all heating power data in each historical record is taken as the heating power characteristic value of each historical record. Here, power is the product of voltage and current, which is a known technique and will not be elaborated further.

[0035] S32: Since the load is a continuous variable, but there may only be a small number of data points for each specific load value, this embodiment sets a load interval every 5kN; calculates the average value of the heating power characteristic value of all excellent historical records where the suspension load characteristic value falls within the same load interval, and uses this average value as the theoretical power of the load interval; the median value of all load intervals and the corresponding theoretical power are used as inputs for least squares method to perform quadratic curve fitting, and a fitting curve for load and power is obtained; further, the output result of the obtained fitting curve is limited to the rated power range of the medium frequency heating equipment (…). ), The fitted curve represents the theoretically optimal power output based on the load. This indicates the rated power of the medium-frequency heating equipment. When the load is too low, it leads to a lower fitted value. When the value is negative, the minimum heating power characteristic value of all excellent historical records within the load interval where the load is located is taken as the theoretical optimal power corresponding to the load. Specifically, if there are no excellent historical records meeting the conditions within the load interval where the load is located, the minimum heating power characteristic value of all excellent historical records within adjacent load intervals is taken. Further, before performing curve calculation, the total number of valid load intervals is determined; if the total number of valid intervals is less than 3, quadratic curve fitting is not performed, and the arithmetic mean of the theoretical heating power of all valid intervals is directly used as the subsequently retrieved theoretical optimal power. When the load is too high, causing the fitted value to be negative... Exceeding When, take This represents the theoretical optimal power corresponding to the load. The least squares fitting process is a well-known method and will not be elaborated upon here.

[0036] Step S4: At the start of a new heating cycle, compare the operating condition data at the start of the new heating cycle with the operating condition data in each valid historical record, and set the starting power of the new heating cycle.

[0037] In one embodiment of this application, step S4 specifically includes: S41: The operating data includes ambient temperature, atmospheric humidity, and suspension point load; the vector composed of the absolute values ​​of ambient temperature and atmospheric humidity is used as the environmental parameter vector; when the equivalent temperature of the heating area drops to... A new heating cycle is initiated; the similarity between the environmental parameter vector at the start of the new heating cycle and the environmental parameter vectors at the start of each valid historical record is calculated. Specifically, in this embodiment, each parameter in the two sets of environmental parameter vectors is normalized, and the Euclidean distance between the two sets of environmental parameter vectors after normalization is calculated. The reciprocal of the sum of the constant 1 and the Euclidean distance is used as the similarity. In this embodiment, the maximum-minimum normalization method is used to normalize all parameters of the same type at all times to achieve the normalization of each parameter. The closer the similarity value is to 1, the closer the absolute physical state of the two environments is.

[0038] S42: The difference between the load at the start of the new heating cycle and the load at the start of each valid historical record is processed to eliminate dimensions, and the load difference is obtained. S43: Based on the comprehensive score of each valid historical record, the similarity, and the load difference, determine the weighting factor for each valid historical record; S44: Retrieve the working condition data of the most recent M valid historical records from the storage unit. In this embodiment, M is set to 5. The implementer can adjust it according to the actual situation. The adjustment range of M is (3, ), The number of valid historical records in the storage unit is given; the product of the intermediate frequency heating power at the start time of each valid historical record and the weighting factor is calculated; the sum of the products of these M valid historical records is calculated as the ratio of the sum of the weighting factors, which is used as the starting power of the new heating cycle.

[0039] In a preferred embodiment of this application, the expressions for the weighting factor and the starting power of the new heating cycle are as follows: In the formula, The weight factor for the m-th valid historical record; The overall score for the m-th valid historical record; The similarity between the environmental parameter vectors at the start time of the m-th valid historical record and the start time of the new heating cycle; The suspension point load at the start of a new heating cycle; The suspension load at the start time of the m-th valid historical record; The value is a preset minimum positive number, used to prevent the denominator from being 0. In this embodiment... The value is 0.0001; This is a function to find the maximum value. The starting power for the new heating cycle; M is the number of valid historical records selected; This represents the medium-frequency heating power at the start time of the m-th valid historical record.

[0040] This indicates the load difference between the start time of the new heating cycle and the start time of each valid historical record; Lieutenant General As a denominator, it serves to... The normalization effect.

[0041] In actual heating processes, the initial stage often faces fluctuations in operating conditions (such as changes in ambient temperature). If the power is directly increased from the static state to the theoretical optimal power based on the current load, it may lead to unstable heating or wasted energy. Therefore, this application uses the above method to reasonably estimate an initial power before heating begins, based on historical data of recent successful heating and the current measured operating conditions, in order to achieve a smooth start-up and shorten the power adjustment time.

[0042] The starting power takes into account the scores of similar historical operating conditions, the similarity of environmental parameters, and the differences in load, which enables the heating equipment to start operating at a more reasonable power level at the beginning of heating, avoiding response delays or energy waste caused by starting from zero power or fixed power.

[0043] Step S5: Calculate the temperature rise efficiency at each moment based on the equivalent temperature change of the heating area and the power consumption, and dynamically adjust the power of the new heating cycle by combining the actual power and the theoretical optimal power obtained based on the actual load.

[0044] In one embodiment of this application, the heating power is dynamically adjusted during the heating process of a new heating cycle, specifically including: S51: First, when the heating equipment heats to any time t within a new heating cycle, the average load within 1 second prior to time t is obtained as the representative load at time t. To eliminate mechanical vibration interference. Furthermore, based on this representative load... The representative load is obtained through the above fitting curve. The corresponding theoretical optimal power Let be the representative power. Then, calculate the load-power fit at time t during the new heating cycle. ,in, The power at the moment preceding time t is used to characterize the actual heating power before dynamic adjustment at time t. Specifically, for the first moment of the cycle's initiation, the acquired starting power of the new cycle is used as the actual power at the previous moment. To determine if the power is matched to the current load, the closer the fit is to 1, the better the current operating condition. When the fit is greater than 1, it means that the power needs to be reduced, and when the fit is less than 1, the power needs to be increased.

[0045] S52: For any time b in any heating cycle B, calculate the equivalent temperature of the heating region at that time. Equivalent temperature of the heating region at the start of heating cycle B The difference between them is divided by the total electrical energy consumed during heating cycle B from the start to time b. The real-time temperature rise efficiency at time b is obtained. The expression is: . The larger the value, the more significant the temperature increase achieved by consuming fixed electrical energy, indicating a better heating effect. A physical proportional relationship between the actual temperature rise and the accumulated energy consumption is established through division.

[0046] S53: The average of the real-time temperature rise efficiency at all times when the total cumulative electrical energy consumed in each heating cycle is greater than 0 is taken as the temperature rise efficiency of each heating cycle; in step S3, the load characteristic values ​​of each excellent historical record are obtained, and among all excellent historical records, the load characteristic value and the representative load at time t in the new heating cycle are selected. The closest excellent historical record is used as the representative load, and the heating cycle corresponding to that excellent historical record is recorded as the representative load. The closest heating cycle; real-time temperature rise efficiency based on time t in the new heating cycle. With representative load Temperature rise efficiency of the closest heating cycle The differences between them, combined with the load-power fit The dynamic power adjustment coefficient at time t in the new heating cycle is obtained. .

[0047] In a preferred embodiment of this application, the expression for the dynamic power regulation coefficient is as follows: In the formula, This represents the real-time temperature rise efficiency at time t during the new heating cycle. , These are the first and second gain coefficients, which are set in this application. , It can be adjusted automatically. Among them, when... When less than 1, If the value is positive, it indicates that the power should be increased; A value greater than 1 indicates a need to reduce power. A value less than 1 indicates low efficiency, requiring increased power or a check of the operating conditions. This is the temperature rise efficiency deviation term.

[0048] Furthermore, considering the inherent thermal inertia in the initial heating phase, an energy consumption threshold is set. In this embodiment, the energy consumption threshold is set to 20% of the total electrical energy consumption of the heating cycle corresponding to the most recent valid historical record. Implementers can set the energy consumption threshold according to their actual situation; this application does not impose specific restrictions. If the cumulative electrical energy consumed from the start of heating to time t in a new heating cycle does not exceed the energy consumption threshold, then when calculating the dynamic power adjustment coefficient, the temperature rise efficiency deviation term is not introduced; that is, the dynamic power adjustment coefficient used at this time is... ; If the energy consumption threshold is exceeded, the complete calculation formula is used, i.e., the dynamic power regulation coefficient used at this time is: .

[0049] S54: To prevent over-adjustment, this application further specifies... Add restrictions, ,Right now The range is [0.6, 1.4]. Furthermore, the theoretical optimal power under load at time t in the new heating cycle is... With dynamic power regulation coefficient The product of these two values ​​serves as the dynamically adjusted power at time t in the new heating cycle. Specifically, the power dynamically adjusted at the start of a new heating cycle. , The starting power for the new heating cycle. The dynamic power adjustment coefficient is the value at the start of a new heating cycle.

[0050] Step S6: Perform medium-frequency heating based on the dynamically adjusted power.

[0051] To avoid frequent power adjustments, this application sets a control cycle, the duration of which is shorter than the average duration of historical heating cycles. The power of the medium-frequency heating device is adjusted once after each control cycle; in this embodiment, the control cycle duration is set to 10 seconds. Implementers can set the control cycle duration according to their actual needs; this application does not impose specific restrictions. Furthermore, the above steps complete the adaptive power adjustment during medium-frequency heating. However, when adjusting according to the above steps, it may result in a large power output even when the target temperature is reached, leading to temperature overshoot. Therefore, this application further incorporates temperature safety constraints, assuming the target temperature is... The equivalent temperature of the heating region at time t is denoted as ,when season , This refers to the rated power of the medium-frequency heating equipment. A preset temperature threshold is set. This allows for low-power heating to be maintained, preventing temperature overshoot. The control cycle is adaptively controlled by sending the control signal to the medium-frequency heating power supply. When The heating cycle ends when the heating lasts for more than 10 seconds. Data such as load, power, and equivalent temperature during this heating cycle are recorded and stored in the storage unit.

[0052] Based on the same inventive concept as the above methods, this application also provides a data analysis-based medium-frequency heating energy-saving control system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described data analysis-based medium-frequency heating energy-saving control methods.

[0053] In summary, this application provides a data analysis-based energy-saving control method for medium-frequency heating. This method filters a "candidate set of optimal operating conditions" from massive historical operating data and establishes a fitting function between the suspension load and the optimal heating power. Based on this, combined with intelligent estimation of initial power and dynamic power correction, the heating power consistently matches the actual load demand throughout the process, effectively avoiding power redundancy and energy waste. Simultaneously, in the initial heating stage, a reasonable initial power is intelligently calculated using scores from historical successful cases, environmental similarity, and load differences, avoiding power surges caused by starting from zero or fixed power. Furthermore, a smooth dynamic adjustment coefficient is used during the heating process, making power adjustments gradual and continuous, reducing the electrical stress on the medium-frequency power supply and heating equipment caused by power surges, and helping to extend equipment lifespan. In addition to dynamic power adjustment, a temperature safety constraint mechanism is introduced. When the real-time temperature approaches the target temperature, the system automatically switches the power to a low-power maintenance mode, avoiding the temperature overshoot problem common in traditional control, reducing energy loss due to overheating, and improving the safety of the heating process and the stability of crude oil processing.

Claims

1. A data analysis-based energy-saving control method for medium-frequency heating, characterized in that, The method includes the following steps: Step S1: Take each continuous heating process in the medium-frequency heating of the oil pumping unit as a heating cycle; take the data within a heating cycle as a historical record; Step S2: Based on the equivalent temperature change of the medium-frequency heating area, the power consumption of heating, and the heating time in each heating cycle, filter each valid historical record and the excellent historical records among the valid historical records; Step S3: Construct a fitting curve of load versus power using the pumping unit suspension point load data and medium-frequency heating equipment power data from excellent historical records, and obtain the theoretical optimal power under each load based on the curve; Step S4: At the start of a new heating cycle, compare the operating condition data at the start of the new heating cycle with the operating condition data in each valid historical record, and set the starting power of the new heating cycle. Step S5: Calculate the temperature rise efficiency at each moment based on the equivalent temperature change of the heating area and the power consumption, and dynamically adjust the power of the new heating cycle by combining the actual power and the theoretical optimal power obtained based on the actual load. Step S6: Perform medium-frequency heating based on the dynamically adjusted power.

2. The data analysis-based medium-frequency heating energy-saving control method as described in claim 1, characterized in that, In step S2, the filtering of each valid historical record and the excellent historical records among the valid historical records includes: S21: If the equivalent temperature of the heating area is greater than the preset target temperature at the end of each heating cycle, then the historical records of each heating cycle shall be regarded as valid historical records. S22: Calculate the unit energy consumption temperature rise for each heating cycle based on the equivalent temperature rise and energy consumption of the heating area in each heating cycle; set the theoretical optimal heating time and compare it with the heating time of each heating cycle to calculate the heating time matching degree of each heating cycle; determine the temperature overshoot of each heating cycle based on the difference between the maximum equivalent temperature and the target temperature in each heating cycle; calculate the comprehensive score of each effective historical record based on the unit energy consumption temperature rise, the heating time matching degree, and the temperature overshoot. S23: Select the preset percentage of historical records with the highest overall scores from all valid historical records as excellent historical records.

3. The data analysis-based medium-frequency heating energy-saving control method as described in claim 2, characterized in that, The theoretically optimal heating time mentioned in step S22 is specifically set as the heating time of the heating cycle in which the target temperature is reached at the historically equivalent temperature and the energy consumption is the lowest. The specific process for calculating the heating time matching degree is as follows: calculate the absolute value of the difference between the heating duration of each heating cycle and the theoretical optimal heating duration, calculate the ratio of the absolute value of the difference to the theoretical optimal heating duration, and take the difference between the constant 1 and the ratio as the heating time matching degree; Specifically, the temperature overshoot is defined as follows: if the difference between the maximum equivalent temperature and the target temperature is greater than 0, then the difference is taken as the temperature overshoot; otherwise, the temperature overshoot is set to 0.

4. The data analysis-based medium-frequency heating energy-saving control method as described in claim 2, characterized in that, The comprehensive score is a weighted sum of the negative correlation mapping value of the temperature overshoot, the temperature rise per unit energy consumption, and the matching degree of the heating time.

5. The data analysis-based medium-frequency heating energy-saving control method as described in claim 2, characterized in that, Setting the starting power for the new heating cycle in step S4 includes: S41: The operating data includes ambient temperature, atmospheric humidity, and suspension point load; The vector composed of the absolute values ​​of ambient temperature and atmospheric humidity is used as the environmental parameter vector; the similarity between the environmental parameter vector at the start of the new heating cycle and the environmental parameter vector at the start of each valid historical record is calculated. S42: The difference between the load at the start of the new heating cycle and the load at the start of each valid historical record is processed to eliminate dimensions, and the load difference is obtained. S43: Based on the comprehensive score of each valid historical record, the similarity, and the load difference, determine the weighting factor for each valid historical record; S44: Calculate the product of the intermediate frequency heating power at the start time of each valid historical record and the weighting factor, and calculate the ratio of the sum of the products of all valid historical records to the sum of the weighting factors, as the starting power of the new heating cycle.

6. The data analysis-based medium-frequency heating energy-saving control method as described in claim 5, characterized in that, The calculation process of the weighting factor in step S43 is as follows: The product of the similarity and the comprehensive score of each valid historical record is used as the numerator, the sum of the load difference and the preset minimum positive number is used as the denominator, and the ratio of the numerator to the denominator is used as the weighting factor.

7. The data analysis-based medium-frequency heating energy-saving control method as described in claim 5, characterized in that, The dynamic adjustment of power for the new heating cycle in step S5 includes: S51: When heating to any moment in a new heating cycle, take the average load within 1 second before the moment as the representative load of the moment, input the fitting curve to obtain the corresponding theoretical optimal power, and record it as the representative power; take the ratio of the power of the moment before the moment to the representative power as the load-power fit degree of the moment. S52: Divide the difference between the equivalent temperature of the heating area at each moment in each heating cycle and the equivalent temperature of the heating area at the beginning of the heating cycle by the total electrical energy consumed from the beginning to each moment of the heating cycle to obtain the real-time temperature rise efficiency at each moment in each heating cycle. S53: Obtain the temperature rise efficiency of each heating cycle, and based on the difference between the real-time temperature rise efficiency at any moment in the new heating cycle and the temperature rise efficiency of the heating cycle with the closest load, combine the load-power adaptability to obtain the dynamic power adjustment coefficient at any moment. S54: The product of the theoretical optimal power under the load at any given moment and the dynamic power adjustment coefficient is taken as the dynamically adjusted power at any given moment in the new heating cycle.

8. The data analysis-based medium-frequency heating energy-saving control method as described in claim 7, characterized in that, The expression for the dynamic power regulation coefficient in step S53 is: If the cumulative energy consumption from the start of the new heating cycle to any given time does not exceed the preset energy consumption threshold, then the dynamic power adjustment coefficient at time t in the new heating cycle is... for: ; otherwise, ; In the formula, The load-power fit at time t. Let be the real-time temperature rise efficiency at time t. The temperature rise efficiency is the heating cycle that is closest to the load at time t. , These are the preset first gain coefficient and second gain coefficient, respectively.

9. The data analysis-based medium-frequency heating energy-saving control method as described in claim 1, characterized in that, In step S6, the intermediate frequency heating based on the dynamically adjusted power includes: The power of the medium-frequency heating device is adjusted once every preset control cycle; when the equivalent temperature of the heating area exceeds the preset temperature threshold, the heating power is reduced to a preset multiple of the rated power of the medium-frequency heating device, wherein the preset multiple is greater than 0 and less than 1; when the equivalent temperature of the heating area exceeds the preset target temperature for a longer than a preset duration, heating is stopped.

10. A data analysis-based medium-frequency heating energy-saving control system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-9.

Images