A method and system for power optimization control of a heating device

Abstract

The application relates to the technical field of optimization control of heat supply units, in particular to a heat supply equipment power optimization control method and system. The method comprises the following steps: acquiring time sequence data generated by a heat supply unit in a preset operation period; determining pressure temperature imbalance of the heat supply unit based on main steam pressure data and the main steam temperature data; determining efficiency energy efficiency coupling degree of the heat supply unit based on heat supply steam extraction amount data and back pressure data; determining a prediction time domain adjustment factor based on the pressure temperature imbalance and the efficiency energy efficiency coupling degree; and adjusting a prediction time domain of a next prediction section after the preset operation period based on the prediction time domain adjustment factor. The method can avoid the limitation of single parameter analysis, and dynamically adapt the prediction time domain of MPC to the working condition of the heat supply unit.

Description

Technical Field

[0001] This application relates to the field of optimized control technology for heating units, specifically to a method and system for optimized power control of heating equipment. Background Technology

[0002] With the growth of global energy consumption and increasing heating demand, energy-saving operation and stable heating quality of heating units have become core objectives for the industry. Main steam pressure, as a key parameter affecting the unit's heat rate and dynamically balancing power generation and heating flow, is crucial for ensuring the performance of the heating system. Model predictive control (MPC) algorithms, due to their ability to optimize control based on system models for future dynamics, are increasingly being applied to the regulation of main steam pressure in heating units, aiming to improve energy efficiency while ensuring heating quality.

[0003] In the existing technology, when using MPC-based control, the main steam pressure of the turbine of the heating unit is usually predicted and controlled based on fixed prediction time domain parameters. The prediction and regulation of future operating conditions are achieved by setting the prediction time domain.

[0004] However, in the above methods, the operation of heating units usually involves multiple key parameters. These key parameters are strongly coupled and time-varying. MPC is based only on fixed prediction time-domain parameters, which makes it difficult to accurately capture the dynamic characteristics of the heating unit operation process. This leads to a decrease in the accuracy of model predictive control, resulting in a mismatch between control commands and actual needs, and affecting the stability and energy efficiency of the power control of heating equipment. Summary of the Invention

[0005] To address the technical problem of decreased accuracy in predictive control based solely on fixed predictive time-domain parameters using MPC, this application aims to provide a power optimization control method and system for heating equipment. The specific technical solution adopted is as follows:

[0006] This application provides a method for optimizing power control of a heating device, the method comprising:

[0007] The system acquires time-series data generated by the heating unit in each prediction segment within a preset operating period. This time-series data includes main steam pressure data, main steam temperature data, heating extraction steam volume data, and back pressure data. Based on the main steam pressure and temperature data within each prediction segment, the system determines the pressure-temperature imbalance of the heating unit in each prediction segment. This pressure-temperature imbalance characterizes the degree of imbalance between the saturation state of the main steam pressure and temperature. Based on the heating extraction steam volume and back pressure data within each prediction segment, the system determines the efficiency-energy efficiency coupling degree of the heating unit in each prediction segment. This efficiency-energy efficiency coupling degree characterizes the degree of abnormal coupling between the heating unit's operating mode and energy efficiency status. Based on the pressure-temperature imbalance and the efficiency-energy efficiency coupling degree of each prediction segment, the system determines the prediction time domain adjustment factor for each prediction segment. Based on multiple prediction time domain adjustment factors within the preset operating period, the system adjusts the prediction time domain of the next prediction segment following the preset operating period.

[0008] Optionally, determining the pressure-temperature imbalance of the heating unit in each prediction segment based on the main steam pressure data and the main steam temperature data in each prediction segment includes: determining the main steam pressure peak data and the main steam temperature peak data in the main steam pressure data of the first prediction segment, wherein the first prediction segment is a prediction segment included in the preset operating period; determining the independent dominant instability between the main steam pressure and the main steam temperature in the first prediction segment based on the main steam pressure peak data and the main steam temperature peak data in the first prediction segment; determining the hysteresis difference between the main steam pressure and the main steam temperature in the first prediction segment based on at least one pair of associated peak data in the first prediction segment, wherein a pair of associated peak data is the main steam pressure peak data and the main steam temperature peak data with the same position; and determining the pressure-temperature imbalance of the first prediction segment based on the independent dominant instability and the hysteresis difference.

[0009] Optionally, determining the independent dominant instability between the main steam pressure and main steam temperature within the first prediction segment based on the main steam pressure peak data and main steam temperature peak data within the first prediction segment includes: determining the sum of the standard deviations of the main steam pressure peak data and the main steam temperature peak data within the first prediction segment; determining the sum of the transfer entropies between the main steam pressure data and the main steam temperature data within the first prediction segment; and determining the independent dominant instability of the first prediction segment based on the sum of the standard deviations and the sum of the transfer entropies.

[0010] Optionally, determining the hysteresis difference between the main steam pressure and the main steam temperature within the first prediction segment based on at least one pair of associated peak data within the first prediction segment includes: pairing the main steam pressure peak data and the main steam temperature peak data within the first prediction segment to obtain the at least one pair of associated peak data; determining a first accumulation result and a second accumulation result, wherein the first accumulation result is the accumulation of the absolute values ​​of the differences between the main steam pressure peak data and the main steam temperature peak data in each pair of associated peak data, and the second accumulation result is the accumulation of the time differences between the occurrence time of the main steam pressure peak and the occurrence time of the main steam temperature peak in each pair of associated peak data; and determining the hysteresis difference of the first prediction segment based on the first accumulation result and the second accumulation result.

[0011] Optionally, determining the efficiency-energy coupling degree of the heating unit in each prediction segment based on the heating steam extraction data and back pressure data in each prediction segment includes: performing trend tests on the heating steam extraction data and back pressure data of the first prediction segment to obtain a first trend feature and a second trend feature. The first trend feature is the trend feature of the heating steam extraction data of the first prediction segment, and the second trend feature is the trend feature of the back pressure data of the first prediction segment. The first trend feature includes a first verification statistic and a first trend term sequence, and the second trend feature includes a second verification statistic and a second trend term sequence. Based on the variance of the first verification statistic and the steam extraction data of the first prediction segment, the degree of influence of heating efficiency in the first prediction segment is determined. This degree of influence of heating efficiency is used to characterize the degree to which the operating mode of the heating unit tends to generate electricity. Based on the distance between the first trend term sequence and the second trend term sequence, and the second verification statistic, the energy efficiency anomaly coupling of the first prediction segment is determined. This energy efficiency anomaly coupling is used to characterize the degree of energy efficiency anomaly of the heating unit. Based on the degree of influence of heating efficiency in the first prediction segment and the energy efficiency anomaly coupling of the first prediction segment, the efficiency-energy efficiency coupling degree of the first prediction segment is determined.

[0012] Optionally, the above-mentioned determination of the prediction time-domain adjustment factor for each prediction segment based on the pressure-temperature imbalance and the efficiency-energy efficiency coupling degree of each prediction segment includes: determining the prediction time-domain adjustment factor for the first prediction segment based on the pressure-temperature imbalance of the first prediction segment, the efficiency-energy efficiency coupling degree of the first prediction segment, and a preset comprehensive evaluation algorithm. The preset comprehensive evaluation algorithm is used to integrate the indicators of the two dimensions into a single evaluation factor. The first prediction segment is any prediction period included in the preset operating period.

[0013] Optionally, the aforementioned preset running segment includes the current prediction segment. Adjusting the prediction time domain of the next prediction segment after the preset running segment based on multiple prediction time domain adjustment factors within the preset running segment includes: adjusting the prediction time domain of the next prediction segment based on the prediction time domain adjustment factor of the current prediction segment when the prediction time domain adjustment factor of the current prediction segment is greater than or equal to the segmentation threshold.

[0014] Optionally, the method further includes: determining the segmentation threshold based on the prediction time-domain adjustment factor of prediction segments other than the current prediction segment within the preset runtime period and the maximum inter-class variance method.

[0015] Optionally, adjusting the prediction time domain of the next prediction segment based on the prediction time domain adjustment factor of the current prediction segment includes: obtaining the prediction time domain of the current prediction segment; determining the adjustment magnitude by multiplying the prediction time domain of the current prediction segment by the prediction time domain adjustment factor of the current prediction segment; and determining the prediction time domain of the next prediction segment by the difference between the prediction time domain of the current prediction segment and the adjustment magnitude.

[0016] This application embodiment also provides a power optimization control system for heating equipment. The system includes a data acquisition module, a data analysis module, an adjustment factor determination module, and a prediction time-domain adjustment module. The data acquisition module is used to acquire time-series data generated by the heating unit in each prediction segment during a preset operating period. This time-series data includes main steam pressure data, main steam temperature data, heating extraction steam volume data, and back pressure data. The data analysis module is used to determine the pressure-temperature imbalance of the heating unit in each prediction segment based on the main steam pressure data and the main steam temperature data. This pressure-temperature imbalance characterizes the relationship between the main steam pressure and the main steam temperature. The data analysis module is used to determine the efficiency-energy efficiency coupling degree of the heating unit in each prediction segment based on the heating steam extraction data and back pressure data in each prediction segment. The efficiency-energy efficiency coupling degree is used to characterize the degree of abnormal coupling between the operating mode and energy efficiency state of the heating unit. The adjustment factor determination module is used to determine the prediction time domain adjustment factor for each prediction segment based on the pressure-temperature imbalance and the efficiency-energy efficiency coupling degree of each prediction segment. The prediction time domain adjustment module is used to adjust the prediction time domain of the next prediction segment after the preset operating segment based on multiple prediction time domain adjustment factors within the preset operating segment.

[0017] This application has the following beneficial effects:

[0018] In this embodiment, multiple time-series data of the heating unit are first acquired. The degree of imbalance of the saturation state of the main steam parameters is analyzed by pressure and temperature data in the time-series data. The coupling anomaly between the heating unit's operating mode and energy efficiency is analyzed by heating extraction steam data and back pressure data, avoiding the limitations of single parameter analysis. Then, the two analyzed indicators determine the prediction time domain adjustment factor and adjust the prediction time domain of the next prediction segment. This enables the prediction time domain of the MPC to dynamically adapt to the operating conditions of the heating unit, effectively improving the accuracy of main steam pressure control, and thus optimizing the power stability and heating quality of the heating unit. Attached Figure Description

[0019] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This application provides a flowchart of a power optimization control method for a heating device according to one embodiment.

[0021] Figure 2 This is a flowchart illustrating another method for optimizing the power control of a heating device according to an embodiment of this application.

[0022] Figure 3 This is a flowchart illustrating another method for optimizing the power control of a heating device according to an embodiment of this application.

[0023] Figure 4 This is a structural diagram of a power optimization control system for a heating device provided in one embodiment of this application. Detailed Implementation

[0024] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a power optimization control method and system for heating equipment proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0026] With the continuous increase in global energy consumption and the increasingly severe problem of climate change, energy conservation has become one of the core objectives in the field. In recent years, the rapid development and popularization of technologies such as the Internet of Things, big data, artificial intelligence, and cloud computing have provided a solid technical foundation for the intelligent upgrading of heating systems. By deploying a large number of sensors, real-time and comprehensive perception of the entire chain of operation data of the heating system from source to network to station to user can be achieved, and in-depth analysis, accurate prediction, and optimization decision-making can be carried out to further ensure the quality and stability of heating on the basis of green energy conservation.

[0027] The main steam pressure of a heating unit's turbine directly affects its heat rate, and regulating the main steam pressure can dynamically balance power generation and heating flow to meet heating load and ensure heating demand. Traditional methods of predictive control (MPC) for the main steam pressure of heating unit turbines often adjust the prediction time domain by analyzing changes in a single fixed factor, neglecting the coupled influence of multiple variables such as heating extraction steam rate, back pressure, and main steam temperature on the main steam pressure. This leads to insufficient capture of the mechanistic interaction characteristics during heating unit operation, exacerbating prediction errors during control, reducing the accuracy of control under complex operating conditions, and resulting in wasted heating energy and unstable heating quality.

[0028] To address the technical problem that traditional methods of predictive control of main steam pressure in heating unit turbines using MPC often adjust the prediction time domain by analyzing changes in a single fixed factor, neglecting the coupling effects between multiple variables such as heating extraction steam rate, back pressure, and main steam temperature, resulting in insufficient capture of the mechanistic interaction characteristics during heating unit operation, exacerbating prediction errors in the heating unit control process, and causing energy waste and unstable heating quality, this application provides a power optimization control method and system for heating equipment.

[0029] The specific scheme of the power optimization control method for heating equipment provided in this application is described in detail below with reference to the accompanying drawings.

[0030] Please see Figure 1 The diagram illustrates a flowchart of a power optimization control method for heating equipment provided in one embodiment of this application.

[0031] like Figure 1 As shown, the power optimization control method for the heating equipment includes S101-S105.

[0032] S101. Obtain the time-series data generated by the heating unit in each predicted segment during the preset operating period.

[0033] The time-series data includes main steam pressure data, main steam temperature data, heating steam extraction data, and back pressure data.

[0034] In one alternative implementation, timing data from a preset runtime segment can be obtained, and then the preset runtime segment can be divided into multiple prediction segments to obtain timing data generated by each prediction segment.

[0035] Optionally, a high-temperature pressure transmitter and a sheathed K-type thermocouple can be installed in the pipeline before the main steam valve of the heating unit turbine to obtain main steam pressure and temperature data; a differential pressure flow meter can be installed in the straight section of the extraction pipeline to obtain heating extraction steam volume data; and an absolute pressure transmitter can be installed at the exhaust port of the low-pressure cylinder of the turbine to obtain back pressure data.

[0036] Optionally, the data from each sensor is collected at the same frequency, and all sensor signals are synchronized by a unified clock through a distributed control system (DCS).

[0037] For example, the sampling frequency can be 1 Hz, the preset runtime can be the previous 24 hours of the current time, and the duration of a prediction segment can be 30 minutes.

[0038] In one alternative implementation, to prevent data loss due to external interference during data acquisition and transmission, and to prevent the impact of different dimensions between different data on subsequent analysis, the missing values ​​of the heating unit's parameter data sequence can be filled using cubic spline interpolation based on timestamps, and the dimensions between various parameter data can be unified through Z-Score standardization.

[0039] Optionally, subsequent data analysis, such as formula calculations, can be performed based on the data after Z-Score standardization.

[0040] S102. Based on the main steam pressure data and main steam temperature data in each prediction section, determine the pressure and temperature imbalance of the heating unit in each prediction section.

[0041] Among them, pressure-temperature imbalance is used to characterize the degree of imbalance between the main steam pressure and the main steam temperature in the saturation state.

[0042] It should be understood that the saturated steam curve is based on the steam thermodynamic property relationship preset by the 1997 industrial standard of the International Association for the Properties of Water and Steam. It is used to establish a one-to-one correspondence between the main steam pressure and the main steam temperature under saturated conditions, and serves as a benchmark for judging whether the actual operating state of steam deviates from the saturated state.

[0043] Understandably, during the operation of a heating unit, under ideal saturation conditions, the main steam pressure and main steam temperature exhibit a strong positive correlation through the saturated steam curve; that is, an increase in main steam pressure will lead to a corresponding increase in its saturated temperature. However, in actual operation, the steam is often in a superheated state. At this time, the main steam temperature changes independently of the pressure, disrupting the balance between the two and thus triggering operational risks.

[0044] When the main steam pressure increases while the temperature remains constant, the superheat of the steam decreases, potentially leading to problems such as wet steam scouring and increased stress on components. Conversely, when the main steam pressure decreases while the temperature remains constant, the enthalpy drop of the steam doing work decreases, requiring an increase in steam flow to maintain unit load. This could cause the turbine's last-stage blades to operate under overload. Therefore, the degree of imbalance between the main steam pressure and temperature can be assessed based on their relationship.

[0045] S103. Based on the steam extraction data and back pressure data of heating in each prediction section, determine the efficiency-energy efficiency coupling degree of the heating unit in each prediction section.

[0046] The efficiency-energy efficiency coupling degree is used to characterize the degree of abnormal coupling between the operating mode and energy efficiency status of the heating unit.

[0047] It should be understood that the heating steam extraction rate data directly reflects the unit's operating mode, i.e., the distribution of steam energy between power generation and heating. Back pressure data, on the other hand, characterizes the work capacity and smoothness of the low-pressure cylinder, and is a key indicator for assessing the unit's energy efficiency. During unit operation, there is a close coupling relationship between the heating steam extraction rate and back pressure. For example, when the operating mode favors power generation, the heating steam extraction rate typically exhibits a specific trend, and the back pressure may also change accordingly. By analyzing the relationship between the heating steam extraction rate data and the back pressure data, characteristics of abnormal coordination between the heating unit's operating mode and energy efficiency status can be identified.

[0048] S104. Based on the pressure-temperature imbalance and efficiency-energy coupling degree of each prediction segment, determine the prediction time-domain adjustment factor for each prediction segment.

[0049] It should be understood that the prediction time domain adjustment factor for each prediction segment is used to characterize the degree of reduction in the prediction time domain when the main steam pressure of the heating unit is predicted and controlled by MPC in each prediction segment during the operation of the heating unit.

[0050] Based on the above description, it should be understood that the pressure-temperature imbalance quantifies the degree of deviation from the thermodynamic equilibrium relationship between the main steam pressure and the main steam temperature, while the efficiency-energy coupling degree quantifies the degree of synergistic anomaly between the unit's operating mode and energy efficiency status. By fusing the features extracted from these two indicators from the perspectives of the thermodynamic system and the energy distribution system, respectively, the predictive time-domain adjustment factor can be obtained.

[0051] Understandably, during the operation of a heating unit, the more severe the disruption of the saturated equilibrium between the main steam pressure and temperature, and the more severe the abnormal coupling between the operating mode and energy efficiency status of the heating unit, the smaller the prediction time domain should be when using MPC to predict and control the main steam pressure of the heating unit. This is to avoid the disruption of the saturated equilibrium or the ambiguity of the complex operating conditions of the heating unit, which could lead to mismatch in the prediction model and a decrease in rapid response capability.

[0052] In one alternative implementation, a weighted average can be taken of the pressure-temperature imbalance and the efficiency-energy coupling degree of each prediction segment to obtain the prediction time-domain adjustment factor for each prediction segment.

[0053] In one implementation of this application embodiment, taking the first prediction segment as an example, the prediction time domain adjustment factor of the first prediction segment can be determined based on the pressure-temperature imbalance of the first prediction segment, the efficiency-energy coupling degree of the first prediction segment, and a preset comprehensive evaluation algorithm.

[0054] The preset comprehensive evaluation algorithm is used to integrate the indicators of the two dimensions into a single evaluation factor, and the first prediction segment is any prediction period included in the preset running segment.

[0055] Specifically, the pressure-temperature imbalance and efficiency-energy coupling degree of the first prediction segment are used as inputs to a preset comprehensive evaluation algorithm to obtain a comprehensive evaluation of the first prediction segment, and this comprehensive evaluation is used as the prediction time-domain adjustment factor for the first prediction segment.

[0056] Optionally, the preset comprehensive evaluation algorithm can be the CIM algorithm.

[0057] Optionally, in this CIM algorithm, the calculation weights of pressure-temperature imbalance and efficiency-energy efficiency coupling degree are the same, so as to ensure that the thermodynamic instability characteristics reflected by pressure-temperature imbalance and the abnormal characteristics of operation mode and energy efficiency synergy reflected by efficiency-energy efficiency coupling degree can be balanced and incorporated into the final decision.

[0058] Understandably, by using a pre-set comprehensive evaluation algorithm to integrate pressure-temperature imbalance and efficiency-energy efficiency coupling, two core indicators from different dimensions are transformed into a single predictive time-domain adjustment factor, avoiding the problem of difficulty in directly combining indicators due to their different dimensions. At the same time, this pre-set comprehensive evaluation algorithm can balance the impact of operating conditions in both dimensions, ensuring that the predictive time-domain adjustment factor is not biased towards a single operating condition, but comprehensively reflects the overall state of the heating unit in terms of both parameter imbalance and energy efficiency anomalies. This provides a unified and comprehensive basis for subsequent predictive time-domain adjustments, avoiding decision-making biases caused by adjusting the time domain based on a single indicator.

[0059] S105. Based on multiple prediction time domain adjustment factors within the preset running segment, adjust the prediction time domain of the next prediction segment after the preset running segment.

[0060] It should be understood that the multiple prediction time-domain adjustment factors within the preset runtime period are the prediction time-domain adjustment factors of the multiple prediction segments included in the preset runtime period.

[0061] In one implementation of this application, the multiple prediction time-domain adjustment factors can be assigned different weights according to their distance from the current time. The closer to the current time, the higher the weight. Then, the weighted average value of each prediction time-domain adjustment factor is determined, and the weighted average value is used as a direct scaling factor. The product of the prediction time domain of the current prediction segment and the scaling factor is used as the prediction time domain of the next prediction segment.

[0062] In another implementation of the embodiments of this application, the preset runtime segment includes the current prediction segment. If the prediction time domain adjustment factor of the current prediction segment is greater than or equal to the segmentation threshold, the prediction time domain of the next prediction segment is adjusted based on the prediction time domain adjustment factor of the current prediction segment.

[0063] It should be understood that since the current prediction segment is the prediction segment closest to the next prediction segment, the prediction time domain of the next prediction segment can be adjusted based solely on the prediction time domain adjustment factor of the current prediction segment.

[0064] It is understandable that if the prediction time domain adjustment factor of the current prediction segment is greater than or equal to the segmentation threshold, it indicates that the prediction time domain adjustment factor of the current prediction segment is large, the saturation imbalance between the main steam pressure and temperature has a greater impact on the efficiency of the heating unit, and the more severe the saturation imbalance of the main steam pressure and temperature in the current prediction segment, the more obvious the energy efficiency anomaly of the heating unit. At this time, it is necessary to adjust the prediction time domain of the next prediction segment to enhance the rapid response capability of the control system.

[0065] If the prediction time domain adjustment factor of the current prediction segment is less than the segmentation threshold, it indicates that the saturation imbalance between the main steam pressure and the main steam temperature in the current prediction segment is relatively minor, and the energy efficiency anomaly of the heating unit is more ambiguous. Therefore, there is no need to adjust the prediction time domain of the next prediction segment.

[0066] In one alternative implementation, the segmentation threshold can be determined based on the prediction time-domain adjustment factor of prediction segments other than the current prediction segment within a preset runtime period and the maximum inter-class variance method.

[0067] Specifically, several candidate thresholds can be generated first to divide the data into two categories: "stationary" and "abnormal". Then, the inter-class variance corresponding to each candidate threshold can be calculated, and the candidate threshold with the largest inter-class variance can be determined as the splitting threshold.

[0068] Understandably, the Otsu's method can automatically find the optimal threshold to distinguish between "adjustment required" and "no adjustment required" operating conditions based on the distribution characteristics of multiple prediction time-domain adjustment factors, thus avoiding the subjectivity and experience dependence of manually setting thresholds.

[0069] It should be understood that since the current prediction segment is adjacent to the next prediction segment, the prediction time domain of the next prediction segment can be determined by referring to the prediction time domain of the current prediction segment and combining it with the prediction time domain adjustment factor of the current prediction segment.

[0070] In one optional implementation, the prediction time domain of the next prediction segment is adjusted based on the prediction time domain adjustment factor of the current prediction segment. Specifically, this can be done by: obtaining the prediction time domain of the current prediction segment; multiplying the prediction time domain of the current prediction segment by the prediction time domain adjustment factor of the current prediction segment to determine the adjustment magnitude; and determining the difference between the prediction time domain of the current prediction segment and the adjustment magnitude as the prediction time domain of the next prediction segment.

[0071] Optionally, if each prediction segment has a predefined initial prediction time domain, the product of the initial prediction time domain and the prediction time domain adjustment factor of the current prediction segment can be used to determine the adjustment magnitude.

[0072] For example, the initial prediction time domain can be 40 minutes.

[0073] It should be understood that the larger the adjustment factor, the more complex the working conditions and the more severe the imbalance and anomalies. In this case, the larger the adjustment range, the faster it can adapt to complex working conditions. When the adjustment factor is small, the adjustment range is also correspondingly small, avoiding control instability caused by excessive adjustment range.

[0074] Optionally, after determining the prediction time domain of the next prediction segment, the prediction time domain of the next prediction segment and the main steam pressure data of the heating unit in the current prediction segment can be used as inputs, and MPC can be used to predict and control the main steam pressure of the heating unit in the next prediction segment.

[0075] In this embodiment, multiple time-series data of the heating unit are first acquired. The degree of imbalance of the saturation state of the main steam parameters is analyzed by pressure and temperature data in the time-series data. The coupling anomaly between the heating unit's operating mode and energy efficiency is analyzed by heating extraction steam data and back pressure data, avoiding the limitations of single parameter analysis. Then, the two analyzed indicators determine the prediction time domain adjustment factor and adjust the prediction time domain of the next prediction segment. This enables the prediction time domain of the MPC to dynamically adapt to the operating conditions of the heating unit, effectively improving the accuracy of main steam pressure control, and thus optimizing the power stability and heating quality of the heating unit.

[0076] Combination Figure 1 ,like Figure 2 As shown, in one implementation of this application embodiment, taking the first prediction segment as an example, the above S102 can be specifically implemented through S201-S204.

[0077] S201. Determine the main steam pressure peak data in the main steam pressure data within the first prediction section, and the main steam temperature peak data in the main steam temperature data within the first prediction section.

[0078] It should be understood that automatic multiscale-based peak detection (AMPD) can adaptively identify local extrema in a data sequence.

[0079] In one alternative implementation, the main steam pressure data and main steam temperature data of the heating units in each prediction segment can be used as inputs, and all peak data in the main steam pressure data and main steam temperature data can be obtained through AMPD.

[0080] S202. Based on the peak data of main steam pressure and main steam temperature in the first prediction section, determine the independent dominant instability between main steam pressure and main steam temperature in the first prediction section.

[0081] It should be understood that when the saturation equilibrium between the main steam pressure and the main steam temperature is broken during the operation of the heating unit, two types of significant changes will occur: on the one hand, the instability of pressure and temperature fluctuation data will increase, and due to the breakdown of the pressure-temperature correspondence, the two data will exhibit a significant unidirectional dominance; on the other hand, after the steam enters an undersaturated or overheated state, due to the influence of the latent heat of vaporization absorption process, the temperature peak value will lag significantly behind the pressure peak value, and the difference between the two peak values ​​will increase.

[0082] Understandably, the independent dominant instability is used to characterize the severity of the fluctuations in the main steam pressure and main steam temperature within a preset prediction range, as well as the degree of unidirectional dominance of their interaction.

[0083] In one alternative implementation, the sum of the standard deviations of the main steam pressure peak data and the main steam temperature peak data within the first prediction segment can be determined first; and the sum of the transfer entropy between the main steam pressure data and the main steam temperature data within the first prediction segment can be determined; then, based on the sum of the standard deviations and the sum of the transfer entropy, the independent dominant instability of the first prediction segment can be determined.

[0084] It should be understood that standard deviation is an indicator used to measure the dispersion of data. The larger the sum of standard deviations, the stronger the volatility of the peak data. Transfer entropy is an indicator used to measure the asymmetry of information flow between two time series. The larger the sum of transfer entropy, the stronger the unidirectional dominance between pressure and temperature changes (i.e., the stronger the predictive power of one series on the other).

[0085] Optionally, the independent dominant instability of a predicted segment satisfies the following formula:

[0086]

[0087] in, Indicates the first Independent dominant instability of each predicted segment, Indicates the first The sum of the standard deviations of all peak data in the main steam pressure and temperature data within each prediction segment. Indicates the first The sum of the transfer entropy between the main steam pressure and temperature data within each prediction segment.

[0088] Optionally, the parameters in the above formula can be normalized.

[0089] Understandably, the sum of standard deviations can intuitively reflect the intensity of fluctuations in the main steam pressure peak data and temperature peak data, while the sum of transfer entropy can accurately capture the unidirectional dominant relationship between the two changes. The independent dominant instability obtained by combining the two covers both the "intensity" of data fluctuations and the "direction" of parameter correlation. This indicator can truly and comprehensively depict the independent instability state of pressure and temperature when saturation imbalance occurs.

[0090] S203. Based on at least one pair of associated peak data within the first prediction segment, determine the hysteresis difference between the main steam pressure and the main steam temperature within the first prediction segment.

[0091] Among them, a pair of related peak data consists of the main steam pressure peak data and the main steam temperature peak data with the same position sequence.

[0092] It should be understood that the hysteresis difference is used to characterize the magnitude of the numerical difference between the main steam pressure peak and the main steam temperature peak within a prediction segment, as well as the degree of hysteresis in the occurrence time of the main steam temperature peak relative to the main steam pressure peak.

[0093] In one alternative implementation, the main steam pressure peak data and the main steam temperature peak data within the first prediction segment can be paired to obtain at least one pair of associated peak data; then, the first accumulation result and the second accumulation result are determined, and based on the first accumulation result and the second accumulation result, the hysteresis difference of the first prediction segment is determined.

[0094] The first cumulative result is the sum of the absolute values ​​of the differences between the main steam pressure peak data and the main steam temperature peak data in each pair of associated peak data; the second cumulative result is the sum of the time differences between the occurrence time of the main steam pressure peak and the occurrence time of the main steam temperature peak in each pair of associated peak data.

[0095] Specifically, the main steam pressure data and main steam temperature data in the first prediction segment can be arranged in time sequence to obtain the main steam pressure data sequence and the main steam temperature data sequence. The peak data with the same position in the main steam pressure data sequence and the main steam temperature sequence are respectively recorded as a pair of associated peak data.

[0096] It is understandable that the main steam pressure peak data is the peak value of the main steam pressure peak, and the main steam temperature peak data is the peak value of the main steam temperature peak.

[0097] It should be understood that the larger the first cumulative result, the greater the difference between the main steam pressure peak data and the main steam temperature peak data, which violates the one-to-one correspondence between pressure and temperature in the saturated steam curve and is direct evidence of imbalance. Therefore, the greater the lag difference in the first prediction segment.

[0098] It should be understood that under normal circumstances, saturated steam needs to absorb latent heat of vaporization to rise in temperature. Therefore, the main steam pressure data will rise to its peak faster than the main steam temperature data. In terms of time, the peak time of the main steam pressure data is earlier than the peak time of the main steam temperature data. The more severe the lag difference, the negative the time difference between the peak times of the main steam pressure data and the main steam temperature data within each pair of correlated peak data. The smaller this negative number is, the smaller the second cumulative result is, indicating that the lag time of the main steam temperature peak relative to the main steam pressure peak is longer, and the greater the lag difference in the first prediction segment.

[0099] Optionally, the lag difference of a prediction segment satisfies the following formula:

[0100]

[0101] in, Indicates the first The degree of lag difference in each prediction segment Indicates the first The first cumulative result is the sum of the absolute values ​​of the differences between the main steam pressure peak data and the main steam temperature peak data in each pair of associated peak data in each prediction segment. Indicates the first The second cumulative result is the sum of the time differences between the occurrence times of the main steam pressure peak and the main steam temperature peak in each pair of correlated peak data within each prediction segment. This represents an exponential function with the natural constant as its base.

[0102] Based on the above formula, it should be understood that, under normal circumstances, the peak time of the main steam pressure data is earlier than the peak time of the main steam temperature data, i.e. A negative value indicates that, in the presence of lag differences, the temperature peak data appears later. It is also a negative number, therefore in real-world scenarios, There are no positive numbers in this formula, with base e. Adding a minus sign in front allows you to achieve... The smaller the value, The larger.

[0103] Optionally, the parameters in the above formula can be normalized.

[0104] Understandably, by pairing peak data with the same positional order, the analysis focuses on the correspondence between pressure and temperature peaks, closely aligning with the core scenario of abnormal parameter peak correlation during saturation imbalance. Since the sum of the absolute values ​​of the differences can quantify the degree of difference in peak size, it directly reflects the imbalance in parameter peak matching. The sum of the time differences can quantify the lag time of the temperature peak relative to the pressure peak, reflecting the inertial influence of vapor phase changes during saturation imbalance. Therefore, the combination of these two methods ensures that the calculation of lag difference accurately covers the two key dimensions of "peak size difference" and "peak time lag," improving the authenticity and accuracy of pressure and temperature imbalance.

[0105] S204. Based on the independent dominant instability and hysteresis difference of the first prediction section, determine the pressure-temperature imbalance of the first prediction section.

[0106] It should be understood that during the operation of a heating unit, the more severe the imbalance between the saturation state of the main steam pressure and the main steam temperature, the stronger the instability of the main steam pressure and temperature fluctuation data of the heating unit turbine, and the higher the unidirectional dominance between the changes in main steam pressure and the changes in main steam temperature. At the same time, the main steam temperature peak generated by the inertial influence of steam phase changes shows a more obvious lag compared to the main steam pressure peak, and the greater the difference between the peak data.

[0107] Optionally, the product of the independent dominant instability and the hysteresis difference of the first prediction segment can be determined, and the result can be normalized to obtain the pressure-temperature imbalance of the first prediction segment.

[0108] Optionally, the pressure-temperature imbalance of a predicted section satisfies the following formula:

[0109]

[0110] in, Indicates the first Pressure-temperature imbalance in each predicted section Indicates the first Independent dominant instability of each predicted segment, Indicates the first The degree of lag difference in each prediction segment This represents the normalization function.

[0111] Understandable, The closer a value is to 1, the more severely the balance between pressure and temperature is disrupted. The closer the value is to 0, the closer the dynamic relationship between pressure and temperature is to the ideal saturation curve.

[0112] In this embodiment of the application, the peak data of main steam pressure and temperature are first extracted using the methods provided in S201-S204 above, focusing on the key characteristics of parameter fluctuations; then the independent dominant instability and hysteresis difference are calculated respectively, and the pressure-temperature imbalance is obtained by combining the two. This method considers both the independent instability characteristics of parameter fluctuations and the hysteresis and difference characteristics of temperature relative to pressure, avoiding fuzzy judgments on the degree of saturation imbalance, and providing accurate and reliable parameter support for the subsequent determination of the time-domain adjustment factor.

[0113] It should be understood that during the operation of a heating unit, when the primary function of the unit is power generation rather than heating, and the unit's energy efficiency anomalies are more severe, the continuous downward trend in the data sequence of the extracted steam for heating becomes more pronounced due to the unit's operating strategy favoring power generation and prioritizing steam for the entire turbine process. The unit's extraction valve remains in a throttling state for extended periods, leading to reduced fluctuations in the extracted steam volume. Simultaneously, the increased low-pressure cylinder exhaust due to increased power generation load results in back pressure higher than normal heating conditions, meaning the unit's back pressure data sequence exhibits a cumulative increase. The inverse coupling between reduced extraction steam and increased back pressure becomes more pronounced. In this situation, it is crucial to reduce the prediction time domain size in the MPC predictive control to prevent the long prediction time domain from becoming blurred during unit operation and to enhance the MPC's rapid response to complex operating conditions.

[0114] Based on the above analysis, an efficiency-energy efficiency coupling state is constructed. In one implementation of this application embodiment, the first prediction segment is still taken as an example, combined with... Figure 1 ,like Figure 3 As shown, the above S103 can be implemented through S301-S304.

[0115] S301. Perform trend tests on the heating steam extraction data and back pressure data of the first prediction section to obtain the first trend feature and the second trend feature.

[0116] The first trend feature is the trend feature of the heating steam extraction volume data of the first prediction section, and the second trend feature is the trend feature of the back pressure data of the first prediction section. The first trend feature includes a first verification statistic and a first trend term sequence, and the second trend feature includes a second verification statistic and a second trend term sequence.

[0117] It should be understood that the check statistic is used to characterize the direction and strength of the data's trend. When the check statistic is negative, it indicates that the data series is in a downward trend, and the smaller the negative value, the more significant the downward trend. When the check statistic is positive, it indicates that the data series is in an upward trend, and the larger the positive value, the more significant the upward trend. The trend term series characterizes the long-term trend of the data series.

[0118] Optionally, the Mann-Kendall trend test algorithm is used to perform trend tests on the heating steam extraction data and the back pressure data of the first prediction segment, respectively, to obtain the first verification statistic and the second verification statistic; the seasonal and trend decomposition using loess (STL) algorithm is used to perform trend tests on the heating steam extraction data and the back pressure data of the first prediction segment, respectively, to obtain the first trend term sequence and the second trend term sequence.

[0119] S302. Based on the variance of the first verification statistic and the steam extraction data of the first prediction section, determine the degree of influence of the heating efficiency of the first prediction section.

[0120] Among them, the degree of impact of heating efficiency is used to characterize the extent to which the operating mode of the heating unit tends to generate electricity.

[0121] In one alternative implementation, since the variance may be 0, a small positive number (i.e., a smoothing factor) can be set. First, the sum of the variance of the heating steam extraction data of the first prediction segment and the smoothing factor is determined. Then, the product between the sum of the variance of the heating steam extraction data of the first prediction segment and the smoothing factor and the first verification statistic is determined. Finally, the absolute value of the product is determined as the degree of influence of the heating efficiency of the first prediction segment.

[0122] It should be understood that the variance of the heating steam extraction data in the first prediction section is used to characterize the fluctuation of the heating steam extraction. The product of the variance and the first verification statistic reflects the continuous low-fluctuation decline in heating steam extraction caused by the shift of the heating unit's operating mode towards the power generation side.

[0123] Optionally, the product result can be normalized so that its value is between 0 and 1.

[0124] S303. Based on the distance between the first trend term sequence and the second trend term sequence, and the second verification statistic, determine the energy efficiency anomaly coupling of the first prediction segment.

[0125] Among them, the energy efficiency anomaly coupling is used to characterize the degree of energy efficiency anomaly of the heating unit.

[0126] It should be understood that the greater the coupling of the energy efficiency anomaly, the greater the degree of energy efficiency anomaly of the heating unit, and the higher the degree of coupling between the energy efficiency anomaly and the change in the operating mode.

[0127] It should be understood that the distance between the first trend term sequence and the second trend term sequence is the dynamic time warping (DTW) distance. The DTW distance between the first trend term sequence and the second trend term sequence can be calculated based on a preset dynamic time warping algorithm. The DTW distance reflects the degree of reverse coupling between the first trend term sequence and the second trend term sequence.

[0128] Optionally, the optimal planning path between the two sequences (i.e., the first trend term sequence and the second trend term sequence) can be determined based on dynamic programming, and the cumulative distance of the optimal planning path can be determined as the DTW distance.

[0129] In one alternative implementation, the ratio of the second verification statistic to the DTW distance can be used to determine the energy efficiency anomaly coupling.

[0130] Alternatively, 1 can be added to the denominator to prevent the DTW from being 0.

[0131] Optionally, the energy efficiency anomaly coupling of a predicted segment satisfies the following formula:

[0132]

[0133] in, Indicates the first Energy efficiency anomaly coupling of each predicted segment, Indicates the first The validation statistics of back pressure data for each predicted segment. Indicates the first The DTW distance between the trend term sequence of the heating steam extraction data and the trend term sequence of the back pressure data within a prediction segment.

[0134] Based on the above formula, it should be understood that, The smaller, The larger the value, meaning the two sequences have similar trend patterns and strong back pressure changes, the better. The value reaches its maximum.

[0135] Optionally, the parameters of the formula can be normalized to eliminate the influence of dimensions.

[0136] Based on the above formula, it should be understood that, The larger the value (i.e., the more significant the upward trend in back pressure), the more inefficient the low-pressure cylinder exhaust is, and the more abnormal the energy efficiency. The larger the value, the greater the difference in shape between the trend of the heating steam extraction data and the trend of the back pressure data, and the stronger the reverse coupling (i.e., the more serious the asynchronous phenomenon of one decreasing and the other increasing). At this time, the energy efficiency anomaly coupling should be stronger.

[0137] S304. Based on the degree of influence of the heating efficiency of the first prediction section and the energy efficiency anomaly coupling of the first prediction section, determine the efficiency-energy efficiency coupling degree of the first prediction section.

[0138] Optionally, the product of the degree of influence of heating efficiency and the coupling degree of energy efficiency anomaly can be determined as the efficiency-energy efficiency coupling degree.

[0139] It should be understood that the greater the efficiency-energy efficiency coupling degree, the more severe the energy efficiency deterioration of the heating unit while its operating mode is tilted towards the power generation side. Therefore, it is more necessary to adjust the prediction time domain of the model predictive control to enhance the rapid response capability of the control system.

[0140] Optionally, the efficiency-energy efficiency coupling degree of a prediction segment satisfies the following formula:

[0141]

[0142] in, Indicates the first The efficiency-energy efficiency coupling degree of each prediction segment Indicates the first The degree of impact on heating efficiency in each predicted section Indicates the first Energy efficiency anomaly coupling of each predicted segment, This represents the normalization function.

[0143] Based on the above formula, it should be understood that the efficiency-energy efficiency coupling degree is simultaneously affected by the degree of influence of heating efficiency and the coupling degree of energy efficiency anomalies. Due to the "amplification effect" of the multiplicative relationship, according to the above formula, the efficiency-energy efficiency coupling degree will only reach a high value when both dimensions show significant anomalies, which is consistent with the meaning of "coupling status".

[0144] Understandable, The closer the value is to 0, the more stable the unit's operating mode is in efficient heating mode, and the better its energy efficiency. The system's operational risk is low. The closer the value is to 1, the more it indicates that the operating mode is biased towards power generation and causes significant energy efficiency anomalies, and the higher the operating risk of the heating unit.

[0145] In this embodiment of the application, trend tests are performed on the heating steam extraction data and back pressure data to obtain verification statistics and trend terms. Since the verification statistics can accurately characterize the direction and intensity of the change trend of the data sequence, and the trend terms characterize the long-term change trend of the data sequence, the analysis of these two indicators can accurately and comprehensively quantify the inverse coupling relationship between the downward trend of heating steam extraction and the upward trend of back pressure.

[0146] like Figure 4As shown in the figure, this application embodiment also provides a power optimization control system for heating equipment. The power optimization control system 40 for heating equipment includes a data acquisition module 401, a data analysis module 402, an adjustment factor determination module 403, and a prediction time domain adjustment module 404.

[0147] The data acquisition module 401 is used to acquire the time-series data generated by the heating unit in each predicted segment during the preset operating period. The time-series data includes main steam pressure data, main steam temperature data, heating steam extraction data, and back pressure data.

[0148] The data analysis module 402 is used to determine the pressure-temperature imbalance of the heating unit in each prediction segment based on the main steam pressure data and the main steam temperature data in each prediction segment. The pressure-temperature imbalance is used to characterize the degree of imbalance between the saturation state of the main steam pressure and the main steam temperature.

[0149] The data analysis module 402 is used to determine the efficiency-energy efficiency coupling degree of the heating unit in each prediction segment based on the heating steam extraction data and back pressure data in each prediction segment. The efficiency-energy efficiency coupling degree is used to characterize the degree of abnormal coupling between the operating mode and energy efficiency status of the heating unit.

[0150] The adjustment factor determination module 403 is used to determine the prediction time-domain adjustment factor for each prediction segment based on the pressure-temperature imbalance and the efficiency-energy coupling degree of each prediction segment.

[0151] The prediction time domain adjustment module 404 is used to adjust the prediction time domain of the next prediction segment after the preset running segment based on multiple prediction time domain adjustment factors within the preset running segment.

[0152] It should be noted that the heating equipment power optimization control system 40 can execute any of the above-mentioned heating equipment power optimization control methods.

[0153] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0154] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for optimizing power control of heating equipment, characterized in that, The method includes: Acquire time-series data generated by the heating unit in each predicted segment during the preset operating period. The time-series data includes main steam pressure data, main steam temperature data, heating steam extraction data, and back pressure data. Based on the main steam pressure data and the main steam temperature data in each prediction segment, the pressure-temperature imbalance of the heating unit in each prediction segment is determined. The pressure-temperature imbalance is used to characterize the degree of imbalance between the saturation state of the main steam pressure and the main steam temperature. Based on the steam extraction data and back pressure data in each prediction segment, the efficiency-energy efficiency coupling degree of the heating unit in each prediction segment is determined. The efficiency-energy efficiency coupling degree is used to characterize the degree of abnormal coupling between the operating mode and energy efficiency status of the heating unit. Based on the pressure-temperature imbalance and the efficiency-energy coupling degree of each prediction segment, the prediction time-domain adjustment factor of each prediction segment is determined. Based on multiple prediction time domain adjustment factors within the preset runtime period, the prediction time domain of the next prediction segment after the preset runtime period is adjusted.

2. The power optimization control method for heating equipment according to claim 1, characterized in that, The determination of the pressure-temperature imbalance of the heating unit in each predicted segment based on the main steam pressure data and the main steam temperature data in each predicted segment includes: Determine the main steam pressure peak data in the main steam pressure data within the first prediction segment, and the main steam temperature peak data in the main steam temperature data within the first prediction segment, wherein the first prediction segment is the prediction segment included in the preset operating period. Based on the peak data of main steam pressure and main steam temperature in the first prediction section, the independent dominant instability between main steam pressure and main steam temperature in the first prediction section is determined. Based on at least one pair of associated peak data within the first prediction segment, the hysteresis difference between the main steam pressure and the main steam temperature within the first prediction segment is determined. The pair of associated peak data consists of the main steam pressure peak data and the main steam temperature peak data with the same position. Based on the independent dominant instability and hysteresis difference of the first prediction segment, the pressure-temperature imbalance of the first prediction segment is determined.

3. The power optimization control method for heating equipment according to claim 2, characterized in that, The determination of the independent dominant instability between the main steam pressure and main steam temperature within the first prediction segment, based on the main steam pressure peak data and main steam temperature peak data within the first prediction segment, includes: Determine the sum of the standard deviations of the main steam pressure peak data and the main steam temperature peak data within the first prediction section; Determine the sum of the transfer entropy between the main steam pressure data and the main steam temperature data within the first prediction segment; The independent dominant instability of the first prediction segment is determined based on the sum of the standard deviations and the sum of the transition entropy.

4. The power optimization control method for heating equipment according to claim 2, characterized in that, The determination of the hysteresis difference between the main steam pressure and the main steam temperature within the first prediction segment, based on at least one pair of associated peak data within the first prediction segment, includes: Pair the main steam pressure peak data and the main steam temperature peak data within the first prediction segment to obtain at least one pair of associated peak data. Determine the first accumulation result and the second accumulation result. The first accumulation result is the sum of the absolute values ​​of the differences between the main steam pressure peak data and the main steam temperature peak data in each pair of associated peak data. The second accumulation result is the sum of the time differences between the occurrence time of the main steam pressure peak and the occurrence time of the main steam temperature peak in each pair of associated peak data. Based on the first and second accumulation results, the lag difference of the first prediction segment is determined.

5. The power optimization control method for heating equipment according to claim 1, characterized in that, The determination of the efficiency-energy coupling degree of the heating unit in each prediction segment based on the heating steam extraction data and back pressure data in each prediction segment includes: Trend tests are performed on the heating steam extraction volume data and back pressure data of the first prediction section to obtain a first trend feature and a second trend feature. The first trend feature is the trend feature of the heating steam extraction volume data of the first prediction section, and the second trend feature is the trend feature of the back pressure data of the first prediction section. The first trend feature includes a first verification statistic and a first trend term sequence, and the second trend feature includes a second verification statistic and a second trend term sequence. Based on the variance of the first verification statistic and the steam extraction data of the first prediction section, the degree of influence of the heating efficiency of the first prediction section is determined. The degree of influence of the heating efficiency is used to characterize the degree to which the operating mode of the heating unit tends to generate electricity. Based on the distance between the first trend term sequence and the second trend term sequence, and the second verification statistic, the energy efficiency anomaly coupling of the first prediction segment is determined, and the energy efficiency anomaly coupling is used to characterize the degree of energy efficiency anomaly of the heating unit. Based on the degree of influence of the heating efficiency of the first prediction section and the energy efficiency anomaly coupling of the first prediction section, the efficiency-energy efficiency coupling degree of the first prediction section is determined.

6. The power optimization control method for heating equipment according to claim 1, characterized in that, The determination of the prediction time-domain adjustment factor for each prediction segment based on the pressure-temperature imbalance and the efficiency-energy efficiency coupling degree of each prediction segment includes: Based on the pressure-temperature imbalance of the first prediction segment, the efficiency-energy coupling degree of the first prediction segment, and the preset comprehensive evaluation algorithm, the prediction time-domain adjustment factor of the first prediction segment is determined. The preset comprehensive evaluation algorithm is used to integrate the two-dimensional indicators into a single evaluation factor. The first prediction segment is any prediction period included in the preset running period.

7. The power optimization control method for heating equipment according to claim 1, characterized in that, The preset runtime segment includes the current prediction segment. Adjusting the prediction time domain of the next prediction segment following the preset runtime segment based on multiple prediction time domain adjustment factors within the preset runtime segment includes: If the prediction time domain adjustment factor of the current prediction segment is greater than or equal to the segmentation threshold, the prediction time domain of the next prediction segment is adjusted based on the prediction time domain adjustment factor of the current prediction segment.

8. The power optimization control method for heating equipment according to claim 7, characterized in that, The method further includes: The segmentation threshold is determined based on the prediction time-domain adjustment factor of the prediction segments other than the current prediction segment within the preset runtime period and the maximum inter-class variance method.

9. The power optimization control method for heating equipment according to claim 7, characterized in that, The step of adjusting the prediction time domain of the next prediction segment based on the prediction time domain adjustment factor of the current prediction segment includes: Obtain the prediction time domain of the current prediction segment; The adjustment range is determined by multiplying the prediction time domain of the current prediction segment with the prediction time domain adjustment factor of the current prediction segment. The difference between the prediction time domain of the current prediction segment and the adjustment magnitude is determined as the prediction time domain of the next prediction segment.

10. A power optimization control system for heating equipment, characterized in that, The system includes a data acquisition module, a data analysis module, an adjustment factor determination module, and a prediction time domain adjustment module; The data acquisition module is used to acquire time-series data generated by the heating unit in each predicted segment during the preset operating period. The time-series data includes main steam pressure data, main steam temperature data, heating steam extraction data, and back pressure data. The data analysis module is used to determine the pressure-temperature imbalance of the heating unit in each prediction segment based on the main steam pressure data and the main steam temperature data in each prediction segment. The pressure-temperature imbalance is used to characterize the degree of imbalance between the saturation state of the main steam pressure and the main steam temperature. The data analysis module is used to determine the efficiency-energy efficiency coupling degree of the heating unit in each prediction segment based on the heating steam extraction data and back pressure data in each prediction segment. The efficiency-energy efficiency coupling degree is used to characterize the degree of abnormal coupling between the operating mode and energy efficiency status of the heating unit. The adjustment factor determination module is used to determine the prediction time-domain adjustment factor for each prediction segment based on the pressure-temperature imbalance and the efficiency-energy coupling degree of each prediction segment. The prediction time domain adjustment module is used to adjust the prediction time domain of the next prediction segment after the preset running segment based on multiple prediction time domain adjustment factors within the preset running segment.

Images