Double-source waste heat mode switching control method oriented to load self-adaption

Abstract

The invention discloses a load-oriented self-adaption double-source waste heat mode switching control method, particularly relates to the technical field of industrial process control and waste heat comprehensive utilization, and aims to solve the problems that heat supply mode switching in an existing double-source waste heat system depends on artificial experience, waste heat supply cannot be matched in a self-adaption mode along with load change, and the working efficiency is low. The waste heat utilization rate is low, the heated load temperature fluctuation is large, and the operation stability is poor. A heat exchange topology model is established in a double-source waste heat system, a load heat demand track and a waste heat capacity fingerprint are synchronously constructed in a rolling time window, and a heat supply mode sequence is optimally selected by taking a load heat demand satisfaction degree, a waste heat utilization degree and a mode switching cost as indexes under an equipment operation constraint and a mode switching constraint. Therefore, the effects of automatically matching the heat load and the waste heat supply under the condition of dynamic change of the working condition and the load, giving consideration to temperature stability and efficient utilization of the waste heat, reducing manual adjustment and improving the system energy efficiency and the operation stability are achieved.

Description

Technical Field

[0001] This invention relates to the field of industrial process control and waste heat utilization technology, specifically a load-adaptive dual-source waste heat mode switching control method. Background Technology

[0002] Existing dual-source waste heat recovery systems, such as units that combine turbine exhaust waste heat and flue gas waste heat, or devices that simultaneously recover process wastewater and flue gas waste heat, generally use local controllers or distributed control systems to logically control each heat exchange unit, circulating pump, and valve. Common practices often involve switching between a few modes, such as "single waste heat source heating" or "dual-path parallel heating," based on fixed priorities or simple temperature, pressure, and load thresholds. Alternatively, operators may manually start / stop pumps and adjust valve openings based on experience, load changes, and field meter readings. These control methods typically lack a unified model of the overall heat exchange relationship between the waste heat source and the heated load, making it difficult to simultaneously characterize changes in waste heat capacity and load demand on the same time scale. Equipment operating boundaries and mode switching frequencies are often constrained by built-in interlocks, simple limits, or shift experience. Control rules are scattered across multiple loops and screens, lacking a systematic quantitative basis.

[0003] In situations with significant fluctuations in operating conditions and loads, such as when the flow rate and inlet temperature of heated media like desulfurization slurry and process circulating water frequently change with upstream process conditions, and when turbine load, furnace combustion level, or production level are also dynamically adjusted, existing technologies often cannot analyze the corresponding load heat demand for each time slot in the future and the sustainable heat exchange capacity of each waste heat source under safe operating conditions within a unified rolling time window. Furthermore, they rarely consider operational constraints such as the inlet and outlet temperature difference of heat exchange units, mixing node temperature, allowable load inlet temperature, pump start-up and shutdown frequency, and valve opening changes, in conjunction with switching constraints such as mode switching sequence and minimum holding time. Therefore, under dynamic operating conditions, the selection and switching of heating modes rely more on instantaneous measurement values ​​and operational experience. This can easily lead to situations where there is surplus waste heat capacity for a certain period but it is not switched in time, or a high load is maintained even when waste heat capacity is insufficient. This results in large fluctuations in the temperature of the heated load, low waste heat utilization, and even problems such as actuator fatigue and process disturbances caused by excessively frequent switching.

[0004] In summary, the current situation in dual-source waste heat systems lacks a control method that can simultaneously quantify changes in load-side heat demand and waste heat supply capacity on a unified time scale, and comprehensively evaluate and select the heating mode sequence while meeting equipment operation constraints and mode switching constraints. Consequently, it is impossible to achieve adaptive and coordinated matching of heating modes to load heat demand and dual-source waste heat supply capacity under dynamic changes in operating conditions and load. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a load-adaptive dual-source waste heat mode switching control method to solve the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a load-adaptive dual-source waste heat mode switching control method, characterized in that it is executed by a control system, the control system including a control device and a heating mode decision module and an execution control module deployed on the control device, comprising: S1. Establish a heat exchange topology model for the dual-source waste heat and the heated load. The heat exchange topology model shall include at least the identification information of the waste heat source, heat exchange unit, mixing node, measurement point, actuator valve and circulation pump, as well as the upstream and downstream connection relationship and control relationship between the waste heat source, heat exchange unit, mixing node, measurement point, actuator valve and circulation pump. S2. Based on the load-side flow rate, load inlet temperature, load outlet temperature, and process target temperature, combined with the allowable heating rate and allowable cooling rate, determine the heat demand for each time slot within the rolling time window, and form a load heat demand trajectory in the order of the time slots. S3. Collect the temperature and flow rate of each waste heat source medium and the upstream load status. Combine the heat exchange capacity estimation model with the temperature and flow rate of the waste heat source medium and the parameters of the heat exchange unit as inputs to determine the heat exchange power and duration that each time segment can provide and construct the corresponding waste heat capacity fingerprint. S4. Based on the heat exchange topology model, load heat demand trajectory and waste heat capacity fingerprint, and superimposed equipment operation constraints and mode switching constraints, a mode feasible domain is generated for the preset waste heat heating mode. S5. Within the rolling time window, the heating mode sequence that meets the mode feasible region is optimized and selected based on the deviation between the actual available heat and the load heat demand trajectory under the candidate heating mode sequence, the utilization ratio of the available heat exchange power in the waste heat capacity fingerprint of the candidate heating mode sequence, and the number of mode switching and valve and circulation pump actions corresponding to the candidate heating mode sequence, so as to determine the target heating mode at the current moment. S6. Establish a mode switching session for the target heating mode, perform mode switching according to the preset sequence of valve and circulating pump actions in the session, monitor the operating deviation of key measurement points, and correct the waste heat capacity fingerprint parameters of subsequent windows based on the deviation.

[0007] Furthermore, S1 includes: The control device will designate the branch that continuously discharges media at a temperature higher than the ambient temperature under normal operating conditions as the waste heat source, the heat exchanger installed between the waste heat source and the heated load as the heat exchange unit, the pipe fittings that realize the confluence and diversion of fluids as the mixing node, and the valves and circulating pumps with electric actuators as the actuators. The control device assigns unique identifiers to waste heat sources, heat exchange units, hybrid nodes, measurement points, and actuators according to a unified coding rule, and generates a heat exchange topology model configuration file that includes object categories, upstream and downstream relationships, associated measurement point identifiers, and associated actuator identifiers.

[0008] Furthermore, S2 includes: The control device is equipped with load-side flow measurement points and load-side temperature measurement points on the load inlet pipeline and the load outlet pipeline; The control device receives the process target temperature, allowable heating rate and allowable cooling rate from the process management system, divides the rolling time window into time slices, and summarizes the load-side flow rate, load inlet temperature and load outlet temperature for each time slice. After outlier removal and sequence smoothing of the data for each time slice, the required heat is calculated, and the load heat demand trajectory is formed in the order of the time slices and stored in the running storage area.

[0009] Furthermore, S3 includes: The control system maps the waste heat source to the heat exchange unit, measurement point and upstream load state according to the heat exchange topology model; The time slices are divided according to the sampling period. Outlier removal and smoothing are performed on the waste heat source temperature and waste heat source flow rate within the time slice to obtain representative temperature and representative flow rate. Based on the heat exchange capacity estimation model, the available heat exchange power for each time slice is estimated using representative temperature and representative flow rate. The duration, host efficiency impact weight, and equipment lifespan impact weight are determined by combining the upstream load status. A parameter set is generated and stored in the operation storage area as a waste heat capacity fingerprint associated with the waste heat source identifier and version identifier.

[0010] Furthermore, S4 includes: The control system establishes equipment operation constraints for heat exchange units, mixing nodes, load inlets, circulating pumps, and actuator valves based on equipment manuals, process procedures, and safety regulations, and establishes mode switching constraints for preset waste heat supply modes. Read the heat demand value of the load heat demand trajectory and the available heat exchange power and duration of the waste heat capacity fingerprint by time slice within the rolling time window; Based on the heat exchange topology model and mode configuration, each preset waste heat heating mode is sequentially checked for heating capacity, temperature and temperature difference boundary, and frequency of circulation pump start-up and shutdown and change of valve opening in each time slot. The combination of the time slot that passes all checks and the preset waste heat heating mode is recorded as the feasible state in the mode feasible domain.

[0011] Furthermore, if the control system determines within the rolling time window that all preset waste heat heating modes have not been recorded as feasible states in the mode feasible domain throughout the entire time slice, it will maintain the current heating mode unchanged. The valve opening and circulating pump operating level corresponding to the current heating mode are limited to a preset safe state, and alarm information indicating that the mode feasible field is empty is output through the human-machine interface and the upper-level scheduling system.

[0012] Furthermore, S5 includes: Within the rolling time window, the heating mode decision module selects preset waste heat heating modes for each time slot based on the mode feasible region, and generates a sequence of candidate heating modes in the order of the time slots. By calling the load heat demand trajectory and waste heat capacity fingerprint according to the time slice, the heat demand deviation, waste heat utilization ratio and mode switching cost of the candidate heating mode sequence in each time slice are calculated. Candidate heating mode sequences that do not meet the preset heat demand threshold and preset waste heat utilization threshold are eliminated. The remaining candidate heating mode sequences are calculated with a comprehensive evaluation score according to the preset evaluation weight. The candidate heating mode sequence with the highest comprehensive evaluation score is determined as the target heating mode in the current time slice, corresponding to the preset waste heat heating mode.

[0013] Furthermore, after determining the target heating mode, the heating mode decision module sends the target heating mode and its corresponding heating mode sequence identifier to the execution control module via the fieldbus. The execution control module verifies valve action commands and circulating pump action commands based on equipment operation constraints and real-time status. When it detects that the target heating mode does not meet the mode feasible domain, the constraint version identifier changes, or the action command exceeds the safe range, it returns feedback information containing error codes and execution version identifiers to the heating mode decision module. The heating mode decision module reselects the heating mode sequence based on the feedback information, and synchronizes the heating mode sequence and version identifier when it detects that the execution version identifier and the decision version identifier are inconsistent.

[0014] Furthermore, S6 includes: When the target heating mode is selected, the control system establishes a mode switching session with a session identifier, and generates valve execution steps and circulation pump steps with action sequence numbers based on the heating mode sequence. The execution control module executes each step according to the action sequence number. After each action is completed, it judges whether the operation deviation exceeds the limit based on the key measurement points and equipment operation constraints. If the deviation exceeds the limit, a safety backoff strategy is executed. After the mode switching session is completed, the heating mode decision module adjusts the waste heat capacity fingerprint parameters based on the deviation between the actual heat exchange and the predicted waste heat capacity fingerprint value.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By establishing a heat exchange topology model in the dual-source waste heat system, including waste heat sources, heat exchange units, mixing nodes, measurement points, and actuators, the load heat demand trajectory and the waste heat capacity fingerprint of each waste heat source are constructed synchronously within a rolling time window. Equipment operation constraints and mode switching constraints are superimposed to form a mode feasible region. Then, the heating mode sequence is optimized and selected using the degree of load heat demand satisfaction, waste heat utilization degree, and mode switching cost as evaluation indicators. This achieves the effect of automatically matching load heat demand and dual-source waste heat supply capacity under dynamic changes in operating conditions and load, prioritizing the selection of target heating modes that take into account both load temperature stability and efficient waste heat utilization, reducing reliance on manual experience adjustment, and improving the overall energy efficiency and operational stability of the dual-source waste heat system.

[0016] 2. By establishing a switching session with a unique session identifier and action sequence number for the target heating mode, valve and pump actions are broken down into detailed steps to be executed sequentially. After each action, the system determines whether to continue execution based on the operational deviation of key measurement points and the preset safety backoff strategy. In long-term operation, the fingerprint parameters are version-corrected based on the deviation between the actual heat exchange and the fingerprint prediction value of the waste heat capacity, and a traceable evidence chain is formed. This achieves the following effects: ensuring the consistency and idempotency of the mode switching process; quickly returning to a safe heating state in abnormal situations; continuously reducing the deviation between the model and the actual operating conditions; and improving the safety, interpretability, and reliability of the entire control strategy. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a load-adaptive dual-source waste heat mode switching control method according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example: Figure 1A flowchart illustrating a load-adaptive dual-source waste heat mode switching control method according to the present invention is provided. This load-adaptive dual-source waste heat mode switching control method is executed by a control system, which includes a control device and a heating mode decision module and an execution control module deployed on the control device, comprising: S1. Establish a heat exchange topology model between the dual-source waste heat and the heated load. The heat exchange topology model should at least include the identification information of the waste heat source, heat exchange unit, mixing node, measurement point, actuator valve, and circulating pump, as well as the upstream and downstream connection relationships and control relationships between the above objects. The specific implementation is as follows: During the construction and renovation of the actual production line, for the dual-source waste heat system, process engineers and automation engineers first jointly identify all equipment and pipelines related to waste heat collection and load heating. Based on the process flow diagram, piping and instrumentation diagram, and equipment layout diagram, the branches in the turbine exhaust cooling circuit, flue gas heat exchange circuit, or process waste hot water circulation circuit that continuously discharge media with temperatures higher than the ambient temperature under normal operating conditions are defined as waste heat sources. The ambient temperature is preferably the sliding average of the outdoor temperature measurement points in the area where the unit is located over one hour as a reference. After the unit has been started up and running stably for at least two hours, the medium temperature and flow rate of each candidate branch are continuously observed. The curve of the quantity changes with the upstream host load is used to gradually adjust the host load from a low level to a high level and then back to the middle level. The corresponding medium temperature is recorded in at least three different load level segments. The sampling window can be set to five consecutive sampling cycles. When the temperature of the branch increases point by point in the sampling window after each load increase and decreases point by point in the sampling window after each load decrease, it can be determined that the temperature change trend of the branch is consistent with the change trend of the host load. The medium temperature is always higher than the ambient temperature throughout the observation period. It can be set to be more than five degrees Celsius higher than the ambient temperature. The branch records that meet the above trend consistency conditions are usable waste heat sources.

[0020] Plate heat exchangers, shell-and-tube heat exchangers, and tubular heat exchangers, which are arranged between the waste heat source and the heated load to achieve heat transfer between the hot and cold media without direct mixing of the hot and cold sides, are defined as heat exchange units. When defining heat exchange units, the locations of the hot and cold inlets of each heat exchange unit are determined by referring to the flow direction arrows of the hot and cold media in the process diagram and the heat exchange area, design flow rate, and inlet / outlet pipe orientation marked on the equipment nameplate. Preferably, at least one temperature measurement point and one flow rate measurement point are configured at each of the hot and cold inlets. During the continuous operation of the heat exchange unit, the records of these measurement points are collected within a process shift, and the changes in temperature and flow rate on the hot and cold sides are compared to support the subsequent calculation of the actual heat exchange capacity of the heat exchange unit under different operating conditions.

[0021] A tee, cross, or junction box where two or more fluids merge or a single fluid splits in a pipeline is defined as a mixing node. When determining a mixing node, the changes in upstream and downstream flow and temperature are observed during a period of time before and after the corresponding valve opening and closing status changes through on-site surveys. The valve is gradually opened from closed to a predetermined opening degree, and at least one sampling period is maintained at each valve position to record the flow and temperature fluctuations of the upstream and downstream branches. This confirms the number and connection relationship of the upstream and downstream branches of each mixing node. Preferably, temperature and pressure measurement points are arranged on a straight pipe section after the merging at the mixing node. In actual operation, the changes in temperature and pressure at this location before and after different heating modes are compared to determine whether the mixing effect meets the expectations of the topology model.

[0022] Temperature measuring devices, flow measuring devices, pressure measuring devices, and valve on / off status acquisition devices installed on the aforementioned waste heat sources, heat exchange units, mixing nodes, and pipelines and equipment under heated loads are uniformly defined as measuring points. When configuring measuring points, thermometers, flow meters, and pressure transmitters with ranges covering the expected operating conditions and accuracy levels meeting process control requirements are selected from the instrument selection manual. A uniform sampling period is set for each type of measuring point in the control device. The sampling period can be set between two and thirty seconds, preferably ten seconds. During the commissioning phase, the signals from these measuring points are continuously collected within a complete process shift. The collected time series is compared with the load change time and environmental conditions recorded in the process log. Each item's change is compared. Records that remain fixed throughout the observation period and show a significant inconsistency with the trend of adjacent measurement points changing with operating conditions, and that have not changed at all within at least sixty consecutive sampling periods, can be marked as frozen records. Significant inconsistency can be manifested as adjacent measurement points experiencing a continuous increase in temperature or flow rate while this measurement point remains unchanged or changes in the opposite direction. Records with values ​​exceeding 10% of the maximum allowable value on the equipment nameplate or falling below 5 degrees Celsius below the aforementioned ambient temperature can be marked as suspicious records that significantly deviate from engineering limits. When establishing the topology configuration file later, these frozen and suspicious records will be removed from the samples used for calculation and judgment, and only retained for reference in the operation log.

[0023] Control valves and shut-off valves connected to the pipe sections of each waste heat source, heat exchange unit, and mixing node, and whose opening degree or state is changed by electric actuators, are defined as actuator valves. Circulation pumps that undertake the task of transporting waste heat working fluid or heated medium, and whose start and stop states can be remotely controlled in the control device, are defined as actuator pumps. When organizing actuator valves and actuator pumps, the circuit number, control cabinet location, and on-site installation orientation of each actuator are checked item by item using electrical circuit diagrams and field number plates. During the trial operation phase, single-step switch commands are issued sequentially. Within three to five sampling cycles after each command is issued, the flow and pressure changes of the connected pipe sections are monitored. When the flow and pressure change within the range specified on the equipment nameplate in the same direction as the valve or pump's movement, it can be confirmed that the control range of the actuator is consistent with the design. The consistency with the direction of movement can be reflected in the gradual increase of flow rate within several consecutive sampling cycles after the valve opening is increased, or the gradual increase of pressure within several consecutive sampling cycles after the pump is started.

[0024] After the physical objects are defined, the control system preferably assigns a unique identifier to each type of object according to a unified coding rule. In this embodiment, the control system includes a field control device, a heating mode decision module deployed on the field control device, and an execution control module. The heating mode decision module is used to perform heating mode sequence optimization and target heating mode determination. The execution control module is used to execute valve actions and circulating pump actions during mode switching sessions. The coding rule can be set to include three parts: object category, system region, and sequence number. During the coding process, the correspondence between object category and coding prefix is ​​first established. Then, according to the process flow from top to bottom, sequence numbers are assigned to the waste heat source, heat exchange unit, mixing node, measurement point, and execution valves and pumps. The coding result is recorded in an object list. Subsequently, in the engineering configuration software of the control device, variable entries are created item by item according to the codes in the object list, and the temperature is uniformly recorded in degrees Celsius. The flow rate is standardized to either volumetric flow rate or mass flow rate per hour, and the pressure is standardized to the same dimension. When importing historical records during the commissioning period, for various flow rate values ​​recorded in liters per minute, cubic meters per hour, or tons per hour, they are converted to the agreed dimension according to the conversion relationship between time units and volume or mass units. The conversion ratio and effective time are recorded during the conversion so that engineers can reconstruct the conversion process when reviewing the original records. After the system has been running continuously for a period of time, such as seven consecutive days, the scheduling engineers can export the records of each measurement point during this period from the control device. For each measurement point, the maximum value, minimum value, and the number of times they occur in different value ranges are counted. Records that are more than 10% higher than the allowable limit on the equipment nameplate or more than 5 degrees Celsius lower than the aforementioned ambient temperature are marked as suspicious values. These records are then manually verified in conjunction with maintenance, shutdown, or instrument replacement during this period, ultimately forming a cleaned set of measurement records.

[0025] After completing the alignment and noise reduction, process engineers and automation engineers jointly generate a configuration file for the heat exchange topology model in the control device based on the cleaned object list. This configuration file can be set as a set of parameters recorded line by line. Each line contains a unique object identifier, object category, upstream and downstream relationship in the process flow, associated measurement point identifier, and associated actuator identifier. When generating the configuration file, the connection relationship between the waste heat source and the corresponding heat exchange unit is recorded first, in the order of the medium flow direction in the process flow. Then, the connection relationship between the heat exchange unit and the mixing node, and between the mixing node and the heated load, is recorded. At the same time, a mode configuration record is reserved for each predetermined heating mode. The mode configuration record lists the state combination that each actuator valve and actuator pump should be in under this mode, so that it can be directly called in the subsequent mode switching session. When the configuration file is generated, a version identifier, effective time, compiler, and reviewer are added. The name of the operator is stored in the non-volatile storage medium of the field control device. Preferably, the loading time and the most recent modification time of the configuration file are recorded in the engineering maintenance interface of the control device so that the topology model version and parameter boundaries used at that time can be traced when the system behavior is investigated in the future. In the fieldbus or industrial Ethernet communication configuration, in order to ensure the consistency of the topology model, mode configuration and parameter configuration between different control units, the field names, field lengths and update rhythms of various objects are specified in the communication configuration table. For example, the waste heat source status field, heat exchange unit status field and hybrid node status field can be set to be updated at a rhythm of about ten seconds. The mode configuration field is only updated at the beginning or end of the mode switching session. The modification permission of the topology model and mode configuration is limited to the authorized engineering maintenance account through permission management. During operation, ordinary on-duty personnel cannot directly change the topology configuration on the control device.

[0026] Through the above steps, a heat exchange topology model is established at the actual production line level, which includes dual-source waste heat, heat exchange units, mixing nodes, measurement points, and the relationships between actuators and actuator pumps. This model is then embedded in the control device in the form of a topology configuration file with version identification and clear source records. This provides a unified, reliable, and traceable physical basis for subsequently constructing load heat demand trajectories, estimating waste heat capacity fingerprints, and switching heating modes within a rolling time window.

[0027] S2. Based on the load-side flow rate, load inlet temperature, load outlet temperature, and process target temperature, combined with the allowable heating rate and allowable cooling rate, determine the heat demand for each time slot within the rolling time window, and form a load heat demand trajectory in the order of the time slots. The specific implementation is as follows: To calculate the heat demand for each time slot within a rolling time window based on load-side flow rate, temperature, and process target temperature, and to form a load heat demand trajectory, the load-side range is first defined in the circuit of the heated medium. Generally, two pipeline sections near the inlet and outlet of the process equipment are designated as the load inlet and outlet pipelines. Flow measurement points installed at these two locations provide the load-side flow rate, and temperature measurement points installed at these two locations provide the load-side temperature. The flow measurement points can be electromagnetic flowmeters or vortex flowmeters with a range covering the expected operating conditions and accuracy meeting process requirements. Temperature measurement points can be resistance temperature detectors (RTDs) or thermocouples. A unified setting is then configured in the control device for both load-side flow rate and load-side temperature. A fixed sampling period is established, which can be set between two and thirty seconds, preferably ten seconds, and remains unchanged within a process shift for subsequent statistical analysis. The process target temperature and allowable fluctuation range are configured by the process engineer in the process management system. The target temperature represents the temperature level that the load outlet is expected to maintain in the long term, and the allowable fluctuation range represents the upper and lower limits of the allowable deviation of the load outlet temperature from the target temperature. When the parameters take effect, the process management system sends the target temperature and allowable fluctuation range, along with the parameter version identifier, to the control system. When the control system receives new parameters, it records the parameter version identifier and the effective time, and always includes the corresponding version identifier in subsequent calculations for traceability.

[0028] To obtain reliable load-side flow and temperature records, the control system backtracks a fixed-duration original sampling sequence from the current moment within each control cycle. This fixed duration can be set to several minutes, such as five minutes. During this time, a quality check is performed on each flow and temperature sample value. The quality check includes outlier identification and short-term disturbance suppression. Outlier identification is performed according to the following rules: For load-side temperature, if a sample value is below zero degrees Celsius or more than 10% above the maximum allowable temperature on the equipment nameplate, that sample value is marked as an outlier. For load-side flow, if a sample value is below 10% of the minimum design flow or more than 10% above the maximum design flow, that sample value is marked as an outlier. Simultaneously, the variation range between adjacent sample values ​​is checked. When the flow or temperature at a certain moment differs from the previous sampling cycle... If a significant abrupt change occurs compared to the previous period's value, such as when the change exceeds twice the design normal fluctuation range, the point can be marked as an outlier. After marking the outlier, the control system does not directly use the point for subsequent calculations. Instead, it replaces the point by interpolating from nearby sampling points or by using the average of two normal sampling values ​​before and after it. This avoids the unreasonable impact of a single outlier on the overall heat demand estimation. Short-term interference suppression can be achieved by smoothing several consecutive sampling values. The smoothing process can be set up so that at each sampling moment, the current moment and several sampling values ​​before and after it are combined into a small window. For example, the window length can be selected as five sampling periods. After sorting these five sampling values, the value at the middle position is taken as the effective value at that moment. Alternatively, a simple arithmetic mean of these five sampling values ​​can be taken as the smoothed value at that moment. This reduces the high-frequency fluctuations caused by sensor jitter and short-term pressure pulsation.

[0029] After outlier removal and smoothing, the control system obtains a continuous and cleaned load-side flow and temperature sequence over a period of time. To facilitate the assessment of heat demand in the future, a rolling time window needs to be defined. The rolling time window is the time range used in the calculation of the load heat demand trajectory. This time range covers a fixed length forward from the current control cycle and is used to reflect the current operating conditions and the demand change trend in the short term. The length of the rolling time window can be set between several minutes and several tens of minutes, preferably a compromise between the response time of the waste heat system and the thermal inertia of the process. For example, it can be set to thirty minutes for a slurry system with a large heat capacity and fifteen minutes for a water system with a small heat capacity. After the window length is determined, the rolling time window is divided into several time slices. Each time slice represents a period of time that can be approximately regarded as stable in terms of thermal balance. The length of the time slice can be set to several minutes, for example, five minutes. Then, the thirty-minute rolling time window is divided into six time slices.

[0030] For each time slot, the control system summarizes the smoothed load-side flow rate and load-side temperature records within the corresponding time period. The representative average flow rate for that time slot can be obtained by averaging all flow rate samples within that time slot. The representative inlet and outlet temperatures for that time slot can be obtained by averaging the load inlet and outlet temperature samples within that time slot. Then, the heat demand is estimated by combining the process target temperature and the allowable heating rate and allowable cooling rate. The allowable heating rate represents the maximum rate at which the load outlet temperature can rise per unit time, and the allowable cooling rate represents the maximum rate at which the load outlet temperature can fall per unit time. These two parameters are also configured by the process engineer in the process management system and sent to the control system. The purpose is to avoid applying excessive heating or cooling loads to the load in a short period of time, which could affect the equipment life or process stability.

[0031] In specific estimation, the control system first calculates the maintenance heat requirement for the time slot within the target temperature range. The maintenance heat requirement can be estimated by multiplying the current load flow rate, the temperature difference between the target outlet temperature and the actual inlet temperature, and the specific heat of the medium. That is, it is the heat required to maintain the load outlet temperature above the lower limit of the target temperature within the time slot. If the load outlet temperature in the previous time slot was below the lower limit of the target temperature, a portion of the heat requirement for the current time slot needs to be added for recovery heat to restore the temperature to the target temperature range. This recovery heat can be calculated based on the temperature difference and the allowable heating rate to estimate the reasonable achievable heating range within the current time slot, and the additional heat required is calculated accordingly. Similarly, if the load outlet temperature in the previous time slot was above the upper limit of the target temperature, a portion of the heat requirement for the current time slot can be deducted, or even appropriate cooling can be allowed, to comply with the allowable cooling rate limit. The control system adds up the maintenance heat requirement and the adjustment amounts related to recovery or cooling in each time slot to obtain the total heat requirement value for that time slot.

[0032] The above calculations are performed sequentially on all time slices within the rolling time window in each control cycle. As time progresses, the rolling time window slides forward by one time slice, discarding the data from the earliest time slice and adding an estimate for the last time slice, thus forming a load heat demand trajectory that updates over time. This trajectory records the heat demand values ​​and their temporal order for multiple consecutive time slices. When recording the heat demand value for each time slice, the control system also adds the corresponding time tag, target temperature version identifier, and version identifiers for the allowed heating rate and allowed cooling rate. These records are stored in a structured manner in the operating storage area of ​​the control device. The operating storage area can quickly retrieve the load heat demand trajectory for any time period by indexing the time tag and version identifier, which can be directly referenced for subsequent waste heat capacity fingerprint matching and heating mode sequence optimization.

[0033] In a preferred embodiment, the load side of a desulfurization slurry circuit is defined as the pipe section between the outlet of the circulating slurry pump and the inlet of the absorption tower. The load side flow rate is provided by an electromagnetic flowmeter at the outlet of the circulating slurry pump, the load inlet temperature is provided by a temperature measurement point at the outlet of the circulating slurry tank, and the load outlet temperature is provided by a temperature measurement point at the inlet of the absorption tower. The control system sets the sampling period to a uniform 10-second interval and the rolling time window to 30 minutes, dividing the 30 minutes into six equal-length time slices, each 5 minutes long. The process engineer configures the target temperature of the desulfurization slurry in the process management system to 65 degrees Celsius, the allowable fluctuation range to ±2 degrees Celsius, and the allowable heating rate to not exceed 1 degree Celsius per minute and the allowable cooling rate to not exceed 1 degree Celsius per minute. After the parameters are sent to the control system, the control system backtracks the original sampling sequence of the most recent 5 minutes in each control cycle, classifying temperatures below zero degrees Celsius or more than 10% above the maximum allowable temperature of the slurry system as outlier values, and temperatures below... Outlier points are marked according to the rule that outlier flow rates are 10% or more than 10% of the minimum slurry flow rate or the maximum slurry flow rate. The remaining data is smoothed using a moving average with a window length of five sampling periods. Then, the average flow rate, average inlet temperature, and average outlet temperature are calculated for each five-minute time slice within the rolling time window. Based on the target temperature and the allowable heating rate and allowable cooling rate, the heat demand required to maintain the outlet temperature between 63 and 67 degrees Celsius within that time slice is estimated. The heat demand values ​​of six consecutive time slices are combined in chronological order to form a load heat demand trajectory. When saving this trajectory, the current target temperature version identifier and the heating rate version identifier are added to ensure that the heat demand can be accurately matched with the process settings used at that time when constructing the waste heat capacity fingerprint and optimizing the heating mode sequence. This allows engineers in the field to reconstruct the load heat demand trajectory calculation process on similar dual-source waste heat production lines by following the above steps and obtain the same judgment results and control basis as in this embodiment.

[0034] S3. Collect the temperature and flow rate of each waste heat source medium and the upstream load status. Combine this with the heat exchange capacity estimation model that uses the temperature and flow rate of the waste heat source medium and the parameters of the heat exchange unit as inputs to determine the heat exchange power and duration available for each time slot, and construct the corresponding waste heat capacity fingerprint. The specific implementation is as follows: The heat transfer capacity estimation model in this embodiment refers to a calculation model used to estimate the available heat transfer power and duration of each time slot based on the waste heat source medium temperature, waste heat source medium flow rate, and heat exchange unit structural parameters. This heat transfer capacity estimation model is related to S1. The heat exchange topology models used to describe the connection relationships between objects are independent yet coordinated. To construct a waste heat capacity fingerprint for each waste heat source within a rolling time window, which can be directly used for subsequent mode decisions, the control system first maps each branch identified as a usable waste heat source to its corresponding heat exchange unit, relevant measurement points, and upstream load status based on the aforementioned heat exchange topology model and object list. The waste heat source medium temperature comes from the temperature measurement points arranged at the inlet and outlet of the waste heat source, and the waste heat source medium flow rate comes from the flow measurement points at the inlet and outlet of the waste heat source or on the circulation branch. The upstream load status is defined as the operating quantity that can reflect the working intensity of the main unit that generates waste heat within the same time period, such as the turbine power generation load, boiler combustion level, or process output level. During the commissioning phase, these operating quantities are coded in a unified manner in the upper-level scheduling system and the local control device, so that the control system can obtain the current main unit power generation, combustion level, output level, and the trend of change in the last few minutes by reading the status records of the upper-level scheduling system and the local control device in each control cycle.

[0035] To ensure a unified time reference for subsequent calculations, the control system uses the same sampling period and time label as the load heat demand trajectory for the waste heat source medium temperature, waste heat source medium flow rate, and upstream load status. Within each control cycle, it traces back a time period equal to the length of the rolling time window, dividing this time period into multiple sub-intervals of the same length as the time slice. For the waste heat source medium temperature and flow rate samples in each sub-interval, outlier removal and smoothing are performed in the same way as on the load side. Outlier determination can follow the rules of being below zero degrees Celsius, more than 10% above the maximum allowable temperature on the equipment nameplate, or 10% below the minimum design flow rate, or 10% above the maximum design flow rate. After removing outliers, the representative temperature and flow rate values ​​for that time slice are obtained by a moving average with a window length of five sampling cycles. At the same time, the upstream load status sampling sequence is divided into the same time slice, and the duration ratio of the upstream load at different levels within each time slice, as well as the load values ​​at the beginning and end of the time slice, are statistically analyzed. By comparing the load values ​​and the direction of change, it is determined whether the main unit is in a high load, stable load, or reduced load state within that time slice.

[0036] In this embodiment, the heat exchange model is defined as a calculation rule reflecting the heat exchange capacity achievable by a heat exchange unit under given medium temperature, flow rate, and equipment structural parameters. During engineering implementation, this heat exchange model can be derived from the heat exchange area, heat transfer coefficient, heat exchange tube arrangement, and specific heat of the hot and cold media through heat balance relationships provided during the design phase. Alternatively, after the device is put into operation, several representative operating points can be selected, and the measured hot and cold side temperatures and flow rates can be compared with the actual heat exchange effect. An empirical heat exchange capacity calculation formula that is approximately valid within the engineering scope can be obtained through fitting. In this embodiment, the heat exchange model is not limited to a specific mathematical expression. Instead, the control system is required to estimate the ideal achievable heat exchange power within each time slice using representative temperature and flow rate values, combined with the effective heat exchange area of ​​the heat exchange unit, the designed heat transfer side flow range, and the allowable inlet and outlet temperature difference of the equipment. This estimation process can be performed according to the following steps: First, within the time slice, take the difference between the representative temperature of the waste heat source medium and the representative temperature of the heated medium at the inlet of the corresponding heat exchange unit as... The first step involves comparing the potential temperature difference with the maximum allowable inlet and outlet temperature difference of the heat exchange unit. If the potential temperature difference exceeds the maximum allowable inlet and outlet temperature difference, the maximum allowable inlet and outlet temperature difference is used as the usable temperature difference, and the excess is considered unusable. The second step compares the representative flow rate of the waste heat source medium within the time slice with the design flow rate range of the heat transfer side of the heat exchange unit. If the representative flow rate is lower than the design minimum heat transfer flow rate, the design minimum heat transfer flow rate is used as the calculation basis. If the representative flow rate is higher than the design maximum heat transfer flow rate, the design maximum heat transfer flow rate is used as the calculation basis, thus avoiding overestimation of heat exchange capacity when the flow rate is extremely low or high. The third step combines the effective temperature difference and effective flow rate after the above trimming with the effective heat exchange area of ​​the heat exchange unit and the heat transfer capacity coefficient in the heat exchange model. Through the product relationship, the estimated value of the usable heat exchange power of the waste heat source within the time slice is obtained without violating the equipment temperature difference limit and flow rate limit. This estimated value is recorded in the control system in the form of a usable heat exchange power field, along with the time slice's time tag and the heat exchange model version identifier.

[0037] To estimate the duration for which the heat exchange power can be continuously utilized within each time slot, the control system analyzes the changing trends of the upstream load status and the representative temperature and flow rate of the waste heat source medium within the rolling time window. It can compare the changes in the upstream load status at the end of the current time slot and the beginning of the next time slot within each time slot. When it is found that the upstream load is switching from a high level to a low level and the temperature of the waste heat source medium shows a decreasing trend over two consecutive time slots, the control system can determine that the available heat exchange power may decrease in subsequent time slots. Therefore, in the estimation of the duration corresponding to the current time slot, only the length of the current time slot is conservatively considered as the duration. When the upstream load remains at a high level for multiple consecutive time slots and the temperature and flow rate of the waste heat source medium remain stable, several subsequent time slot lengths can be superimposed on the estimation of the duration corresponding to the current time slot, but the total duration cannot exceed the length of the rolling time window to avoid over-extrapolating long-term trends.

[0038] To reflect the impact of different utilization levels on host efficiency and equipment lifespan, the control system, while calculating the available heat exchange power, statistically analyzes the deviation between the representative temperature of the waste heat source medium and the host's design exhaust steam temperature or flue gas design temperature within each time slice. When the representative temperature is close to the lower or upper limit of the design allowable range, a larger host efficiency impact weighting coefficient can be set, and the heat exchange power obtained under this temperature condition will be marked as the part that may have a significant impact on host efficiency. For cases where the flow rate is close to the boundary of the equipment's allowable range, a larger equipment lifespan impact weighting coefficient can also be set, and the heat exchange conditions operating under this flow rate condition will be marked as the part that may have a significant impact on the equipment's mechanical fatigue. In actual implementation, the weighting coefficient can be mapped to one or several levels by a preset ratio of the deviation between the representative temperature and the design temperature, and then the level can be written into the host efficiency impact field of the corresponding time slice. Similarly, the deviation between the representative flow rate and the boundary of the design heat transfer flow rate range can be mapped to one or several levels by a preset ratio, and then the level can be written into the equipment lifespan impact field of the corresponding time slice.

[0039] Through the above steps, within each rolling time window, the control system can obtain a set of parameters for each waste heat source, including the estimated available heat exchange power, the estimated duration range, the weight level of the main unit efficiency, and the weight level of the equipment lifespan. This set of parameters constitutes the waste heat capacity fingerprint of the waste heat source in the current rolling time window. When generating the waste heat capacity fingerprint, the control system records the unique identifier of the waste heat source, the time of generation of the waste heat capacity fingerprint, the version identifier of the heat exchange model used, the version identifier of the upstream load state parsing rule used, and the length of the rolling time window. The fingerprint is stored in the operation storage area in the form of a structured record. In subsequent heating mode decisions and switching session correction actions, the available heat exchange power and related weight information of the corresponding time slice can be read by matching the waste heat source identifier and time tag.

[0040] In a preferred embodiment, a unit is equipped with a turbine exhaust heat exchange unit and a flue gas heat exchange unit. The turbine's designed exhaust temperature is 50 degrees Celsius, with a design allowable fluctuation range of ±5 degrees Celsius. The flue gas's designed temperature is 150 degrees Celsius, with a design allowable fluctuation range of ±10 degrees Celsius. The control system sets the rolling time window to 30 minutes and the time slice length to 5 minutes. During the unit's high-load phase, the upstream load status record shows that the unit's power generation load is above 90% of its rated power and remains stable over multiple consecutive time slices. At this time, the turbine exhaust temperature fluctuates slowly between 48 and 52 degrees Celsius, and the exhaust flow rate is close to 80% to 90% of the designed heat transfer flow rate. The available heat transfer power estimated by the heat transfer model can be set to a range of 200,000 to 300,000 kilojoules per hour, and the corresponding main unit efficiency influence weight can be mapped to a medium level. When similar results are obtained continuously over multiple time slices within the rolling time window, the control system can record a longer sustainable time period for the turbine exhaust side in the waste heat capacity fingerprint. For the flue gas side heat exchange... In the heat unit, during the higher flue gas temperature stage (between 150°C and 160°C), the flow rate is close to about 70% of the design heat transfer flow rate. The available heat transfer power estimated by the heat transfer model is slightly lower than that of the turbine exhaust side, and can be set to the range of 100,000 to 200,000 kJ per hour. However, since the flue gas temperature is close to the maximum allowable temperature boundary of the equipment, the weight of the equipment lifespan impact is mapped to a higher level. The control system separately marks this part of the available heat transfer power in the waste heat capacity fingerprint. In the subsequent heating mode decision, by comparing the load heat demand trajectory with the above waste heat capacity fingerprint, it can prioritize the use of waste heat from the turbine exhaust side, which has a smaller impact on the main unit efficiency and can be used for a longer duration, and can also assist in the use of waste heat from the flue gas side in a short period of time. This improves the overall waste heat utilization level while ensuring that the system is not in a high-temperature, high-flow-rate condition for a long time. Engineers in this field can reconstruct the same waste heat capacity fingerprint calculation process in a similar dual-source waste heat system by following the above steps of signal acquisition, time slice division, heat transfer capacity estimation, duration evaluation, and weight coefficient mapping.

[0041] S4. Based on the heat exchange topology model, load heat demand trajectory, and waste heat capacity fingerprint, and superimposed with equipment operation constraints and mode switching constraints, a mode feasible region is generated for the preset waste heat heating mode. The specific implementation is as follows: Based on the established heat exchange topology model, load heat demand trajectory, and waste heat capacity fingerprint, the control system generates the feasible domain for each preset waste heat heating mode. First, it organizes equipment operation constraints and mode switching constraints according to equipment manuals, process procedures, and safety regulations. Equipment operation constraints are a unified description of the allowable operating range of each heat exchange unit, mixing node, load inlet, pump, and actuator valve during continuous operation. The upper and lower limits of the inlet and outlet temperature difference for each heat exchange unit are given in degrees Celsius by consulting heat exchanger design documents and operating experience records. The allowable temperature range for the mixing node and the allowable temperature range for the load inlet are also specified. The temperature range is set by process engineers as an upper and lower limit range based on the physical properties of the heated medium, the risk of scaling, and product quality requirements. This range is then configured in the control system as the alarm upper and lower limits for the corresponding measurement points. The pump's allowable start-stop frequency is determined by statistical analysis of the equipment manufacturer's recommended start-stop frequency and operation and maintenance records, and is written into the parameter table in the form of allowable start-stop frequency per hour or per day. The allowable opening range and action interval of the execution valve are determined by consulting valve selection data and interlocking logic. The minimum safe opening degree, maximum long-term opening degree, and minimum time interval between two adjacent actions for each type of valve are written into the constraint configuration of the control device in specific values.

[0042] Mode switching constraints describe the time and sequence restrictions of each preset waste heat heating mode. They include the minimum holding time of each mode, the allowed switching order, and the prohibited switching paths. The minimum holding time can be set between several minutes and several tens of minutes, determined by the process engineer in combination with the thermal inertia of the heated medium and the sensitivity of the site to frequent switching. For example, it can be set to fifteen minutes and recorded in the mode constraint table. The allowed switching order is achieved by listing the target modes that can be directly switched from the current mode. The prohibited switching paths are achieved by listing the combination of modes that are not allowed to be directly jumped to, thereby avoiding large fluctuations in the system caused by crossing multiple states in a short period of time.

[0043] The preset waste heat supply modes are defined in advance based on the combination relationship between waste heat sources and heat exchange units in the heat exchange topology model. For example, they are defined as a single waste heat source supply mode that is only heated by the turbine exhaust side, another waste heat source supply mode that is only heated by the flue gas side, a multi-stage series heating mode in which the turbine exhaust side and the flue gas side are used in stages through multiple heat exchange units, and a parallel heating mode in which two waste heat sources supply heat to the same mixing node in parallel. Each mode is assigned a mode identifier during the configuration phase, and the target state combination of each valve and pump in the configuration file is listed in the configuration file, including which valves should be in the open or closed position, which pumps should be in the running or stopped state, and the corresponding target opening degree or running level.

[0044] After the above constraints and mode definitions are completed, the control system constructs a rolling time window with the current time as a reference within each control cycle, and uses the aforementioned time slice division results as a time grid. It then performs a feasibility check on each preset waste heat heating mode in each time slice. When checking the feasibility of a mode in a certain time slice, the control system first reads the heat demand value from the load heat demand trajectory corresponding to that time slice from the operating storage area, and then reads the estimated available heat exchange power and duration of each waste heat source in that time slice from the corresponding waste heat capacity fingerprint. For the waste heat sources included in the current mode, it then... Their available heat exchange power is combined according to the series or parallel relationship in the topology. For example, in the series mode, the downstream heat exchange unit can be reduced sequentially due to the limitation of the upstream heat exchange capacity. In the parallel mode, the available heat exchange power of the two channels can be summed and compared with the load heat demand value to determine whether the mode can theoretically meet the heat demand in this time slot. When the combined available heat exchange power is lower than a certain preset percentage threshold of the load heat demand value, such as lower than 90%, the control system marks the time slot as insufficient heat in this mode, which is considered as not meeting the heating capacity requirements in the equipment operation constraints.

[0045] If the heat capacity is sufficient, the equipment operation constraints are checked. In this mode, based on the topology model and mode configuration, the possible inlet and outlet temperatures and temperature difference ranges of each heat exchange unit, the possible mixing temperature range of the mixing node, and the expected trend of the load inlet temperature can be calculated within this time slice. The control system compares the available heat exchange power and duration in the waste heat capacity fingerprint with the designed heat transfer capacity of the heat exchange unit to estimate the upper and lower limits of the outlet temperature of each heat exchange unit within this time slice. These estimation results are then compared one by one with the upper and lower limits of the inlet and outlet temperature difference of the heat exchange unit, the allowable temperature range of the mixing node, and the allowable temperature range of the load inlet. When any estimated temperature difference or temperature exceeds the allowable range, this time slice is marked as violating the equipment operation boundary in this mode, thus determining that this time slice is not feasible in this mode.

[0046] After the above checks are passed, the control system further checks whether the pump start-stop frequency and valve opening changes will exceed the limits in the time slice. If the current mode remains the same in the next time slice, it is only necessary to determine whether the pump and valve action counts in the already occurred switching session have not exceeded the allowed number of start-stops and actions within the statistical period. If it is necessary to switch from the current mode to this mode or from this mode to other modes, it is necessary to combine the minimum action interval and minimum hold time recorded in the mode switching constraints, compare the current time slice position with the previous entry and exit time slices of this mode. When the time interval is less than the configured minimum hold time or the time interval between two adjacent actions is less than the configured minimum action interval, the time slice is marked as violating the mode switching constraints and is not feasible in this mode.

[0047] After completing the checks on heat capacity, temperature and temperature difference boundaries, and frequency and time intervals of operation, for a given time slot, if all checks pass under a preset waste heat heating mode, the control system marks the corresponding position of the time slot and mode identifier in the mode feasible domain as feasible; otherwise, it marks it as infeasible. After scanning all time slots and all preset modes within the entire rolling time window, a set of feasible markers arranged by time slot and mode dimensions is formed. This set is the mode feasible domain of the preset waste heat heating mode under the current rolling time window. When saving the mode feasible domain, the control system adds equipment operation constraint version identifiers and mode switching constraint version identifiers so that the specific constraint parameters used when a mode in a given time slot is determined to be infeasible can be traced later.

[0048] When the control system detects that all preset waste heat heating modes are marked as infeasible across all time slots within a certain rolling time window (i.e., the mode feasible domain is empty), the control system, according to the pre-set safety strategy, will not attempt to initiate new mode switching sessions. Instead, it will maintain the current heating mode unchanged and lock the opening or operating level of the corresponding valves and pumps in a safe state. For example, it will close some non-critical valves, keep critical circulation pumps running and limit their operating level from being increased. The system will also issue notification information through the human-machine interface and the upper-level dispatching system, prompting operators to check whether the equipment operating constraints and mode switching constraints need to be adjusted.

[0049] In a preferred embodiment, a dual-source waste heat system sets the rolling time window to 30 minutes and the time slice length to 5 minutes. For the multi-stage series heating mode, the mixing node temperature is specified to be no higher than 75 degrees Celsius. For the parallel heating mode, a preset lower limit is set for the outlet temperature difference of each heat exchange unit to be no less than 10 degrees Celsius. The minimum hold time for each mode is uniformly configured to 15 minutes. In actual operation, when the load heat demand trajectory shows that the heat demand for the next two time slices is large, and the waste heat capacity fingerprint of the turbine exhaust side indicates that its available heat exchange power and duration are sufficient to cover these two time slices, after verifying that the estimated temperature and temperature difference of each heat exchange unit do not exceed the above boundaries, and that the pump start-up and shutdown frequency and valve action interval all meet the constraints, the control system... In the feasible domain of the model, the corresponding positions of these two time slices in the series heating mode are marked as feasible. However, for some heating modes that frequently rely on high-temperature flue gas from the flue gas side in a short period of time, due to the high weight of equipment lifespan and the estimated temperature being close to the upper limit of equipment allowable temperature, the control system marks the relevant time slices as infeasible during the equipment operation constraint check. This allows the series heating mode or the single heating mode on the turbine exhaust side to be selected first from the feasible domain during subsequent mode optimization, thereby reducing the occurrence of high-risk operating conditions while meeting the load heat demand. Engineers in this field can reproduce the same feasible domain construction process in similar dual-source waste heat systems by following the above steps of signal reading, constraint configuration, time slice checking and marking.

[0050] S5. Within the rolling time window, based on the deviation between the actual available heat supply and the load heat demand trajectory under the candidate heating mode sequence, the utilization ratio of the available heat exchange power in the waste heat capacity fingerprint of the candidate heating mode sequence, and the number of mode switching times and the number of valve and circulating pump actions corresponding to the candidate heating mode sequence, the heating mode sequence that meets the mode feasible region is optimized and selected to determine the target heating mode at the current moment. The specific implementation is as follows: Based on the previously generated heat exchange topology model, load heat demand trajectory, waste heat capacity fingerprint, and mode feasible region, the heating mode decision module in the control system optimizes the heating mode sequence around the current rolling time window in each control cycle. In this process, the candidate space is first limited according to the mode feasible region, that is, each time slot can only select the preset waste heat heating mode that is marked as feasible in that time slot. Then, starting from the actual mode state of the time slot in the current control cycle, the heating mode sequence is expanded from front to back in the time dimension. During the expansion process, mode maintenance or switching is only allowed between adjacent time slots in a manner permitted by the mode feasible region and mode switching constraints, thereby generating a set of candidate heating mode sequences that meet the basic constraints.

[0051] To evaluate the merits of candidate heating mode sequences, the decision-making module quantifies three evaluation indicators—load heat demand satisfaction, waste heat utilization, and mode switching cost—by time slice within a rolling time window. Load heat demand satisfaction measures the deviation between the actual available heat capacity under the given heating mode sequence and the heat demand value in the load heat demand trajectory for that time slice. During calculation, the decision-making module, based on the heating mode corresponding to the sequence in each time slice, calls the estimated available heat exchange power of the corresponding waste heat source in the waste heat capacity fingerprint, and combines these values ​​through topological relationships to obtain the optimal heat exchange power for that time slice. The available heat of the sequence is then subtracted from the load heat demand value. The absolute value of the difference is taken as the heat demand deviation for that time slot. The deviations of each time slot are weighted and summed or the maximum value is calculated according to the time slot length. This is then compared with the load heat demand satisfaction threshold configured in the parameter table in advance. For example, it can be set to ensure that the average deviation does not exceed a certain percentage of the heat demand value over the entire rolling time window or that the deviation of any time slot does not exceed a certain percentage of the heat demand value. Only when the deviation of the candidate heating mode sequence remains within the threshold range for all time slots within the rolling time window is the sequence considered to be qualified in terms of load heat demand satisfaction.

[0052] Waste heat utilization is used to measure the utilization of available heat in the waste heat capacity fingerprint by the candidate heating mode sequence. This avoids situations where large amounts of low-cost waste heat are abandoned for extended periods in pursuit of stable load temperatures. During calculation, the decision module compares the actual heat exchange power of each waste heat source in the sequence with the estimated available heat exchange power of each waste heat source in the waste heat capacity fingerprint for each time slice in each time slice. The ratio of actual power utilization to available power is statistically analyzed for each time slice. For example, the average or minimum value of the ratio for each waste heat source within the rolling time window is calculated to reflect the performance of that waste heat source. The system calculates the utilization rate of the heat exchange in the sequence and accumulates the available heat exchange power that is not used in each time slice to obtain the amount of waste heat in the window. Then, it compares the waste heat utilization rate with a pre-set waste heat utilization threshold, such as requiring the average utilization rate to be no less than a certain percentage and the amount of waste heat to not exceed a certain threshold. When a candidate heating mode sequence has a very low utilization rate on a certain waste heat source for a long time or the cumulative amount of waste heat in the entire window is significantly higher than that of other sequences, the decision module can determine that the sequence is unqualified in terms of waste heat utilization and remove it.

[0053] The mode switching cost is used to measure the impact of mode switching behavior in a heating mode sequence on equipment lifespan and process stability. During calculation, the decision module counts the number of mode switches occurring within a rolling time window, the time interval between each switch, and the total number of valve and pump actions involved in each switch. When counting actions, it compares changes in valve opening and pump operating status between adjacent time slices, recording the number of times valves need to be switched from closed to open, from open to closed, or their opening adjusted, pumps started / stopped, and pump speeds changed. These three factors are then weighted according to preset weights to obtain the mode switching cost score for the sequence. The weights can be set during the engineering configuration phase based on the equipment's sensitivity to frequent starts / stops and opening changes, with a higher weight for pump start / stop frequency, a lower weight for valve actions, and a medium weight for switching frequency and intervals based on the process's tolerance to disturbances. Within a rolling time window, the decision module standardizes and weights the above three indicators for each candidate heating mode sequence according to a unified evaluation weight version identifier. The weighted summation yields the comprehensive evaluation score for the sequence. Thresholds for load heat demand satisfaction and waste heat utilization are used as hard constraints. Specifically, candidate heating mode sequences that do not meet the load heat demand satisfaction or waste heat utilization thresholds are eliminated before the comprehensive evaluation. For the remaining candidate heating mode sequences that meet these thresholds, they are sorted based on the comprehensive evaluation score and the mode switching cost score. Generally, heating mode sequences with higher load heat demand satisfaction, higher waste heat utilization, and lower mode switching cost are preferred. When multiple heating mode sequences have similar comprehensive evaluation scores, the decision module can select the one with lower mode switching cost or more stable performance in recent operation as the final adopted heating mode sequence. The preset waste heat heating mode corresponding to the current control cycle time slice is extracted from this heating mode sequence as the target heating mode for the current moment. The target heating mode, the corresponding heating mode sequence identifier, the current rolling time window identifier, and the adopted evaluation weight version identifier are all written into the operation log for traceability.

[0054] Subsequently, the heating mode decision module sends the target heating mode, the mode summary of the corresponding heating mode sequence in several future time slices, and the expected mode holding time to the execution control module via the fieldbus. During the sending process, it is agreed that if the heating mode decision module, while reviewing the mode feasible region in a subsequent control cycle, finds that the current target heating mode no longer meets the mode feasible region in a certain time slice or that the constraint version has changed, affecting the feasibility of the heating mode sequence, it will provide feedback to the execution control module in the form of error code 101 and trigger a new round of decision-making. When the execution control module attempts to execute the valve and pump action sequence under the target heating mode, it will consider equipment operating constraints and... When the equipment determines in real-time that certain actions cannot be performed within the safe range, it will report this to the heating mode decision module in the form of error code 102, prompting the decision module to select another heating mode sequence from the mode feasible domain. When the heating mode decision module and the execution control module find that the currently effective decision version number and execution version number are inconsistent, for example, the execution control module is still executing according to the heating mode sequence of the previous rolling time window while the decision module has switched to a new round of heating mode sequence, the execution control module will mark it in the form of error code 104, prompting both sides to synchronize and correct the currently used heating mode sequence and version identifier.

[0055] In a preferred embodiment, the evaluation weight corresponding to the degree of satisfaction of load heat demand can be set to be significantly greater than the evaluation weight of waste heat utilization and mode switching cost. For example, the weight ratio of the three can be configured as 5:3:2. The decision-making effect under different evaluation weight combinations is compared during the operation of multiple consecutive natural days. By statistically analyzing the time proportion of the load outlet temperature within the target temperature range, the average number of mode switching within the rolling time window, and the changes in the number of start-ups and shutdowns of valves and pumps, the rationality and robustness of the evaluation rules are verified. This guides engineers to adjust the evaluation weights and threshold parameters accordingly in similar dual-source waste heat systems and reproduce the above-mentioned heating mode sequence optimization and target heating mode decision-making process.

[0056] S6. Establish a mode switching session for the target heating mode, perform mode switching according to the preset sequence of valve and circulating pump actions in the session, monitor the operational deviation of key measurement points, and correct the waste heat capacity fingerprint parameters of subsequent windows based on the deviation. Specifically, the implementation is as follows: Once the heating mode decision module selects the target heating mode within a certain control cycle, it first establishes a switching session corresponding to this mode switch in the control system. When the switching session is established, a unique session identifier is generated within the current rolling time window. This session identifier is associated with the current heating mode and the target heating mode one by one. The mode states corresponding to each time slice from the current time slice start point to the expected target mode before stabilization are read from the aforementioned heating mode sequence. The differences in mode states between adjacent time slices are parsed into a series of valve and pump action steps arranged in chronological order. Each step clearly indicates the unique identifier of the valve or pump to be adjusted, the direction of opening change or start / stop status change, and the minimum time interval to wait before the expected action is completed. At the same time, the heating mode sequence identifier, action sequence number start value, expected completion time, and safety rollback strategy to be executed during the switching process are recorded for the entire session. The safety rollback strategy is generally predetermined by the process engineer and operation engineer during the parameter configuration stage. It can be set to roll back to the current heating mode, roll back to a pre-selected safe heating mode, or enter a backup state that only retains the necessary circulating pump operation if a serious parameter deviation or signal loss occurs during the switching process.

[0057] After receiving the session switching information, the execution control module associates the session identifier with the local control queue and reads the valves and pumps that need to be adjusted in each step in ascending order of action sequence number. Before each action step begins, it checks whether the action sequence number of the session is consecutive with the sequence number of the previously completed action. If the action sequence number is found to be backward or there is a missing number in the middle, the execution control module considers it an out-of-order instruction, refuses to execute the action, and writes the session identifier, the abnormal action sequence number and the current time into the operation log. At the same time, it sends back the error information of version inconsistency or abnormal sequence to the heating mode decision module.

[0058] When the action sequence number conforms to the natural increment rule, the execution control module issues an opening adjustment command to the corresponding execution valve or a start / stop command to the corresponding pump. After issuing the execution signal, it pauses for a period of time according to the preset waiting rhythm. This waiting rhythm can be set to several seconds to several tens of seconds in the engineering configuration. The specific value is determined according to the pipeline volume and medium flow rate on site. When the waiting rhythm ends, the execution control module reads the latest sampled values ​​of the mixing node outlet temperature, load inlet temperature, and load loop pressure from the aforementioned key measurement points, as well as the reference values ​​of the previous control cycle before the action. By comparing whether the temperature and pressure change in the expected direction and whether the change range is within the allowable deviation range during this observation period, it is determined whether the current action is successful. The allowable deviation of the key measurement points is set by the process engineer during the parameter configuration stage. Generally, it is defined as a certain degree Celsius plus or minus the target temperature and a certain ratio of the equipment's allowable pressure. In practice, the execution control module compares the load inlet temperature with the target temperature and the allowable fluctuation range in the observation window after each action. When the temperature deviates from the upper or lower limit of the target temperature for several consecutive sampling cycles by more than the allowable deviation, it is determined that the operating deviation exceeds the limit.

[0059] Simultaneously, the outlet temperature of the mixing node and the load loop pressure are compared with the upper limits of temperature, temperature difference, and pressure configured in the equipment operation constraints. When the pressure exceeds a certain proportion of the equipment's allowable value or the temperature approaches the material's allowable limit, it is also considered a serious deviation. In these cases, the execution control module stops subsequent actions according to the safety rollback strategy pre-recorded in the switching session, cancels the already executed actions in reverse order, adjusts them to the safe heating mode or maintains the original heating mode state according to the pre-agreed method, and writes the session identifier, the name of the key measurement point that triggered the rollback, the measured value and allowable range, and error code 1031 into the operation log for subsequent traceability.

[0060] When the deviation has not yet reached a severe level but the direction of deviation is opposite to the expected direction for a short period of time, the execution control module can extend the waiting rhythm in the same session and collect the changes of key measurement points in the observation window again. If the expected trend is still not reached after multiple rounds of observation, a safety backoff is triggered to avoid frequent interruptions of the switching process due to transient fluctuations. To ensure the sequentiality and idempotency of instruction execution, the execution control module maintains a session status table locally for each switching session. The table records the maximum action sequence number that has been successfully executed. When an action instruction is received from the heating mode decision module, if the session identifier is the same and the action sequence number is less than or equal to the maximum sequence number recorded in the session status table, it is considered that the instruction is duplicated with the already executed instruction and will not be executed again. Status confirmation information is only fed back when necessary. The actual execution process is entered only when the action sequence number is greater than the maximum sequence number in the session status table and there are no missing numbers. After the action is completed and no serious deviation occurs in the key measurement points, the maximum sequence number in the session status table is updated to the action sequence number.

[0061] When all actions during the session switching are successfully completed sequentially and key measurement points remain within the allowable fluctuation range for the preset minimum holding time, the heating mode decision module compares the actual heat exchange with the predicted waste heat capacity fingerprint value according to the time slices in the rolling time window in several subsequent control cycles. The actual heat exchange can be estimated by combining the load inlet and outlet temperatures, load-side flow rate, and medium specific heat in each time slice. The decision module keeps this estimation process consistent with the heat calculation formula used in the aforementioned load heat demand trajectory construction stage to ensure that the predicted value and the actual value can be directly compared within the same time slice. If it is found that the deviation between the actual heat exchange and the available heat exchange power in the waste heat capacity fingerprint of a certain waste heat source is consistently high in multiple consecutive rolling time windows, which is manifested as the actual heat exchange being continuously less than the predicted value, the decision module will take action. Furthermore, when the deviation exceeds the preset proportional threshold, the available heat exchange power parameter and sustainable duration parameter of the waste heat source in the waste heat capacity fingerprint are adjusted by a certain downward ratio when the adjustment cycle arrives. For example, the current parameter value is multiplied by an adjustment coefficient less than one, and a new waste heat capacity fingerprint version identifier is assigned to the adjusted parameter. When the deviation is consistently low, which is manifested as the actual heat exchange continuously exceeding the predicted value and the deviation exceeding the threshold on the other side, the above parameters are corrected by a certain upward ratio to improve the degree of fit of the waste heat capacity fingerprint to the actual operating conditions. During each parameter adjustment, the heating mode decision module stores the parameter values ​​before and after the adjustment, the actual heat exchange in the corresponding time period, and the original records of key measurement points in a structured manner in the evidence chain record for subsequent operation evaluation and, when necessary, review of the rationality of parameter adjustment.

[0062] In a multi-module control architecture, when different process units or upper-level scheduling initiate multiple adjustment requests for different target heating modes to the heating mode decision module within a short period of time, the control system sorts the switching sessions according to the session establishment time and puts the session requests into the queue in order. When the queue detects requests that point to the same target heating mode and have the same heating mode sequence within the same time period, these requests can be merged into a single switching session to avoid repeatedly initiating switching operations for the same group of valves and pumps. At the same time, the system configuration sets an upper limit on the number of switching sessions that are currently in execution. When the number of switching sessions that are currently being executed or waiting to be executed reaches this upper limit, subsequent new session requests are postponed or returned to the queue, and the corresponding session identifier and the reason for postponement are recorded.

[0063] In a preferred embodiment, the allowable deviation of key measurement points can be configured as ±2 degrees Celsius for the target temperature and no more than 90% of the equipment's allowable pressure for the load loop. The waiting rhythm in the switching session can be set to 20 to 30 seconds, and the minimum hold time can be set to 15 minutes. In the operating data of multiple consecutive successful switching sessions, the deviation distribution between the actual heat exchange of each waste heat source and the predicted value of the waste heat capacity fingerprint can be statistically analyzed. When it is found that the actual heat exchange of a certain waste heat source has a deviation higher than ±10% in more than half of the more than ten rolling time windows, its available heat exchange power parameter can be adjusted up or down by 5%. In subsequent operation, the deviation distribution can be continuously monitored to see if it converges, so as to gradually correct the waste heat capacity fingerprint to a level that is highly consistent with the actual operating conditions. Engineers in the art can reproduce the same mode switching session control process in similar dual-source waste heat systems by following the above steps of session establishment, action execution, deviation monitoring, safety rollback, parameter tuning, and concurrent management.

[0064] In the operating scenario shown in this embodiment: In a desulfurization system of a coal-fired power unit employing dual waste heat sources from both the turbine exhaust side and the boiler flue gas side, the operating unit, following step S1 during the initial commissioning phase, has process engineers and automation engineers jointly review all equipment and pipelines in the turbine exhaust cooling circuit, boiler flue gas heat exchange circuit, and desulfurization slurry heating circuit. Based on the process flow diagram, piping and instrumentation diagram, and equipment layout diagram, the turbine exhaust cooling water branch and flue gas cooling water branch, which continuously discharge temperatures higher than ambient under normal operating conditions, are designated as waste heat sources. The plate heat exchangers and shell-and-tube heat exchangers located between the waste heat sources and the desulfurization slurry load are designated as heat exchange units. The tees and junctions in the slurry pipeline that perform confluence and diversion functions are designated as mixing nodes. Temperature measuring devices installed on each pipe section are also designated as heat exchange units. Flow measurement devices, pressure measurement devices, and valve on / off status acquisition devices are uniformly calibrated as measurement points. Control valves and shut-off valves that adjust their opening through electric actuators, as well as circulating pumps responsible for conveying slurry and cooling water, are calibrated as actuator valves and circulating pumps. After completing on-site verification and trial operation, automation engineers assign unique identifiers to waste heat sources, heat exchange units, mixing nodes, measurement points, actuator valves, and circulating pumps according to unified coding rules. In the control device engineering configuration software, a heat exchange topology model configuration file is generated, recording the object category, upstream and downstream relationships, associated measurement point identifiers, and associated actuator identifiers. The version identifier and effective time are attached and stored in the non-volatile storage medium of the on-site control device, providing a unified topology basis for subsequent control.

[0065] After the unit enters the stable operation phase, during the daytime shift of a certain natural day, process engineers configure the target temperature, allowable fluctuation range, allowable heating rate, and allowable cooling rate for the desulfurization slurry circuit in the process management system. Following step S2, the control system selects the load inlet pipeline and load outlet pipeline on the pipeline between the desulfurization absorber inlet and the circulating slurry tank outlet, and sets up load-side flow measurement points and load-side temperature measurement points at these two locations respectively, uniformly setting the sampling period to ten seconds. Within each control cycle, the control system traces back the original sampling sequence of load-side flow and load-side temperature from the current moment for the most recent five minutes, performing outlier removal and sequence smoothing to obtain continuous... The load-side flow rate and load-side temperature are recorded. Based on this, the rolling time window is set to thirty minutes and divided into six time slices. In each time slice, the representative flow rate, representative load inlet temperature, and representative load outlet temperature are calculated. Combining the target temperature with the allowable heating rate and allowable cooling rate, the maintenance heat requirement required to keep the load outlet temperature in the target range is calculated. The recovery heat or cooling and load reduction heat is calculated based on the load outlet temperature deviation between adjacent time slices. The total heat requirement value of each time slice is obtained by superimposing the values. The load heat demand trajectory is formed in the order of the time slices and stored in the operation storage area along with the time tag, target temperature version identifier, and heating / cooling rate version identifier.

[0066] Within the same time period, the control system, following step S3, utilizes the heat exchange topology model and object list established in S1 to map the turbine exhaust cooling water branch and boiler flue gas cooling water branch to their corresponding heat exchange units, relevant measurement points, and upstream load status variables. Within each control cycle, using the same sampling period and time label as the load heat demand trajectory, it backtracks the waste heat source medium temperature, waste heat source medium flow rate, and upstream unit load records for the most recent 30 minutes. This time period is divided into sub-intervals with the same time slice length. Outlier removal and smoothing are performed on the waste heat source medium temperature and flow rate within each time slice to obtain representative temperature and flow rate. Simultaneously, the upstream unit power generation load sequence is statistically classified into high load, stable load, and reduced load states within the same time slice. The decision module calls upon the heat exchange model pre-tuned during the design and commissioning phases to map the representative temperature and flow rate to the heat exchange... By combining the unit's effective heat exchange area, the designed heat transfer flow range, and the allowable inlet and outlet temperature difference of the equipment, the estimated available heat exchange power is calculated for each time slice under the condition that the temperature difference and flow rate do not exceed the limits. The duration is estimated based on the upstream load status change trend. While calculating the available heat exchange power, the deviation between the representative temperature and the unit's design exhaust steam temperature and flue gas design temperature is mapped as the host efficiency influence weight, and the deviation between the representative flow rate and the boundary of the design heat transfer flow range is mapped as the equipment lifespan influence weight. Finally, within the current rolling time window, a parameter set is formed for each waste heat source, including the estimated available heat exchange power, the estimated duration range, the host efficiency influence weight level, and the equipment lifespan influence weight level. This parameter set is used as the waste heat capacity fingerprint of the corresponding waste heat source in this window, and is stored in the running storage area along with the waste heat source unique identifier, heat exchange model version identifier, and upstream load status parsing rule version identifier.

[0067] Based on the above, the control system, following step S4, constructs the mode feasibility domain for the predefined single heating mode on the turbine exhaust side, single heating mode on the flue gas side, multi-stage series heating mode on the turbine exhaust side and flue gas side, and parallel heating mode on the turbine exhaust side and flue gas side. Specifically, in implementation, firstly, based on the equipment manual, process regulations, and safety regulations, operational constraints are established for each heat exchange unit, mixing node, load inlet, circulating pump, and actuator valve. The allowable temperature difference range between the inlet and outlet of the heat exchange unit, the allowable temperature range of the mixing node, and the allowable temperature range of the load inlet are determined. The allowable start / stop frequency of the circulating pump and the allowable opening degree and action interval of the actuator valve are configured in the control device. Simultaneously, the minimum hold time, allowable switching sequence, and prohibited switching path are configured for each preset waste heat heating mode in the mode constraint table. Then, within each control cycle, a rolling time window is constructed with the current time as a reference, and the heat demand value in the load heat demand trajectory is indexed according to the time slice. The estimated available heat exchange power and duration in the waste heat capacity fingerprint are used to perform heating capacity checks, temperature and temperature difference boundary checks, and checks on the frequency of circulation pump start-up and shutdown and the opening changes of execution valves for each preset waste heat heating mode in time slice order. When a certain time slice meets the heating capacity, all equipment operation constraints, and mode switching constraints under a certain preset waste heat heating mode, the combination of that time slice and the preset waste heat heating mode is recorded as a feasible state in the mode feasible domain. After scanning all time slices and all preset modes in the rolling time window, a mode feasible domain is formed by arranging them according to the time slice dimension and the mode dimension. If all preset waste heat heating modes are not recorded as feasible states in all time slices within a certain window, the control system keeps the current heating mode unchanged and locks the execution valves and circulation pump operation levels under the current mode in the preset safe state. An alarm message indicating that the mode feasible domain is empty is output through the human-machine interface and the upper-level scheduling system.

[0068] After obtaining the feasible region of modes, the heating mode decision module optimizes the heating mode sequence within the current rolling time window according to step S5. Within a certain control cycle, starting from the actual heating mode of the time slice in the current control cycle, it selects modes for each time slice only from the set of preset waste heat heating modes allowed in the feasible region of modes, and expands from front to back in the time slice dimension to form several sets of candidate heating mode sequences that meet the basic constraints. The decision module calls the load heat demand trajectory and waste heat capacity fingerprint according to the time slice for each candidate heating mode sequence, calculates the heat demand deviation between the available heat and the heat demand value under the sequence for each time slice, and calculates the waste heat utilization ratio by the ratio between the actual heat exchange power called by each waste heat source and the estimated value of the available heat exchange power. At the same time, it counts the number of mode switching, the time interval between adjacent switching, and the changes in the opening degree of the executed valve and the circulation pump. The number of start / stop actions is compared with the preset heat demand satisfaction threshold and the waste heat utilization ratio with the preset waste heat utilization threshold. Candidate heating mode sequences that do not meet the thresholds are eliminated. For the remaining candidate heating mode sequences that meet the thresholds, the decision module standardizes and weights the load heat demand satisfaction, waste heat utilization, and mode switching cost according to the pre-configured evaluation weight version identifier to obtain a comprehensive evaluation score. The candidate heating mode sequences are then sorted according to the comprehensive evaluation score. The preset waste heat heating mode corresponding to the candidate heating mode sequence with the highest comprehensive evaluation score in the current time slice is determined as the target heating mode for the current control cycle. The target heating mode, the corresponding heating mode sequence identifier, the current rolling time window identifier, and the evaluation weight version identifier are recorded in the operation log and sent to the execution control module via the fieldbus.

[0069] In scenarios where the target heating mode requires switching from a single heating mode on the turbine exhaust side to a multi-stage series heating mode, the control system establishes a mode switching session and executes mode switching control according to step S6. After selecting the target heating mode, the heating mode decision module generates a mode switching session with a session identifier in the control system. It extracts the mode state from the current time slice start point to the expected stabilization of the target mode from the aforementioned heating mode sequence. It parses the mode state differences between adjacent time slices into a series of valve execution steps and circulation pump steps arranged in chronological order. Each action step clearly records the unique identifier of the executed valve or circulation pump, the direction of opening change or start / stop status change, and the minimum waiting time interval before the expected action is completed. It also records the corresponding heating mode. The system includes a thermal mode sequence identifier, an action sequence number start value, an expected completion time, and a safety backoff strategy. After receiving the mode switching session, the execution control module adds the session identifier to the local control queue and reads each action step sequentially from smallest to largest action sequence number. Before executing each action step, it checks whether the action sequence number is consecutive to the maximum completed sequence number. If an action sequence number is found to be backwards or missing, the instruction is marked as out-of-order and an error message is returned. When the action sequence number conforms to the natural increment rule, the execution control module issues an opening adjustment instruction to the corresponding execution valve or a start / stop instruction to the corresponding circulating pump. After execution, it monitors the mixing node outlet temperature and load inlet temperature within multiple sampling periods according to a preset waiting rhythm. The system monitors the temperature and load loop pressure, recording key measurement points and comparing them with the upper limits of temperature, temperature difference, and pressure configured in the equipment operation constraints. When pressure exceeds the allowable range or temperature approaches the material's allowable limit, the execution control module stops subsequent actions according to the pre-configured safety rollback strategy in the switching session. It also reverses or adjusts the executed actions to a safe heating mode, recording the session identifier, the name of the key measurement point triggering the rollback, the measured value and allowable range, and the corresponding error code in the operation log. When all action steps are completed sequentially and the key measurement points remain within the allowable fluctuation range for the minimum holding time, the heating mode decision module continuously monitors the load inlet temperature, load outlet temperature, and load loop pressure in subsequent control cycles. Load-side flow is collected, and the actual heat exchange is estimated using the same heat calculation method as that used to construct the load heat demand trajectory. This is then compared with the predicted value of the current version of the waste heat capacity fingerprint. If it is found that the actual heat exchange of a certain waste heat source is consistently lower than the predicted value and the deviation exceeds the preset proportional threshold within multiple consecutive rolling time windows, the available heat exchange power parameter and the duration parameter of that waste heat source in the waste heat capacity fingerprint are lowered when the adjustment cycle arrives. A new waste heat capacity fingerprint version identifier is assigned to the adjusted parameters. The parameter values ​​before and after the adjustment are consistent with the actual heat exchange in the corresponding time period and are stored as evidence in the operation storage area. In subsequent operations, the deviation distribution is continuously monitored until convergence, so that the waste heat capacity fingerprint maintains a high degree of consistency with the actual operating conditions.Engineers in this field can complete the deployment and operation of the dual-source waste heat mode switching control system by following the steps S1 to S6 above in similar dual-source waste heat systems, and obtain the operating effect of high load heat demand matching degree, full utilization of waste heat and small impact on equipment during mode switching.

[0070] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0071] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0072] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0073] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0074] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0075] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0076] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0077] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0079] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A load-adaptive dual-source waste heat mode switching control method, characterized in that, Executed by a control system, which includes a control unit and a heating mode decision module and an execution control module deployed on the control unit, including: S1. Establish a heat exchange topology model for the dual-source waste heat and the heated load. The heat exchange topology model shall include at least the identification information of the waste heat source, heat exchange unit, mixing node, measurement point, actuator valve and circulation pump, as well as the upstream and downstream connection relationship and control relationship between the waste heat source, heat exchange unit, mixing node, measurement point, actuator valve and circulation pump. S2. Based on the load-side flow rate, load inlet temperature, load outlet temperature, and process target temperature, combined with the allowable heating rate and allowable cooling rate, determine the heat demand for each time slot within the rolling time window, and form a load heat demand trajectory in the order of the time slots. S3. Collect the temperature and flow rate of each waste heat source medium and the upstream load status. Combine the heat exchange capacity estimation model with the temperature and flow rate of the waste heat source medium and the parameters of the heat exchange unit as inputs to determine the heat exchange power and duration that each time segment can provide and construct the corresponding waste heat capacity fingerprint. S4. Based on the heat exchange topology model, load heat demand trajectory and waste heat capacity fingerprint, and superimposed equipment operation constraints and mode switching constraints, a mode feasible domain is generated for the preset waste heat heating mode. S5. Within the rolling time window, the heating mode sequence that meets the mode feasible region is optimized and selected based on the deviation between the actual available heat and the load heat demand trajectory under the candidate heating mode sequence, the utilization ratio of the available heat exchange power in the waste heat capacity fingerprint of the candidate heating mode sequence, and the number of mode switching and valve and circulation pump actions corresponding to the candidate heating mode sequence, so as to determine the target heating mode at the current moment. S6. Establish a mode switching session for the target heating mode, perform mode switching according to the preset sequence of valve and circulating pump actions in the session, monitor the operating deviation of key measurement points, and correct the waste heat capacity fingerprint parameters of subsequent windows based on the deviation.

2. The load-adaptive dual-source waste heat mode switching control method according to claim 1, characterized in that, S1 includes: The control device will designate the branch that continuously discharges media at a temperature higher than the ambient temperature under normal operating conditions as the waste heat source, the heat exchanger installed between the waste heat source and the heated load as the heat exchange unit, the pipe fittings that realize the confluence and diversion of fluids as the mixing node, and the valves and circulating pumps with electric actuators as the actuators. The control device assigns unique identifiers to waste heat sources, heat exchange units, hybrid nodes, measurement points, and actuators according to a unified coding rule, and generates a heat exchange topology model configuration file that includes object categories, upstream and downstream relationships, associated measurement point identifiers, and associated actuator identifiers.

3. The load-adaptive dual-source waste heat mode switching control method according to claim 1, characterized in that, S2 include: The control device is equipped with load-side flow measurement points and load-side temperature measurement points on the load inlet pipeline and the load outlet pipeline; The control device receives the process target temperature, allowable heating rate and allowable cooling rate from the process management system, divides the rolling time window into time slices, and summarizes the load-side flow rate, load inlet temperature and load outlet temperature for each time slice. After outlier removal and sequence smoothing of the data for each time slice, the required heat is calculated, and the load heat demand trajectory is formed in the order of the time slices and stored in the running storage area.

4. The load-adaptive dual-source waste heat mode switching control method according to claim 1, characterized in that, S3 include: The control system maps the waste heat source to the heat exchange unit, measurement point and upstream load state according to the heat exchange topology model; The time slices are divided according to the sampling period. Outlier removal and smoothing are performed on the waste heat source temperature and waste heat source flow rate within the time slice to obtain representative temperature and representative flow rate. Based on the heat exchange capacity estimation model, the available heat exchange power for each time slice is estimated using representative temperature and representative flow rate. The duration, host efficiency impact weight, and equipment lifespan impact weight are determined by combining the upstream load status. A parameter set is generated and stored in the operation storage area as a waste heat capacity fingerprint associated with the waste heat source identifier and version identifier.

5. The load-adaptive dual-source waste heat mode switching control method according to claim 1, characterized in that, S4 include: The control system establishes equipment operation constraints for heat exchange units, mixing nodes, load inlets, circulating pumps, and actuator valves based on equipment manuals, process procedures, and safety regulations, and establishes mode switching constraints for preset waste heat supply modes. Read the heat demand value of the load heat demand trajectory and the available heat exchange power and duration of the waste heat capacity fingerprint by time slice within the rolling time window; Based on the heat exchange topology model and mode configuration, each preset waste heat heating mode is sequentially checked for heating capacity, temperature and temperature difference boundary, and frequency of circulation pump start-up and shutdown and change of valve opening in each time slot. The combination of the time slot that passes all checks and the preset waste heat heating mode is recorded as the feasible state in the mode feasible domain.

6. The load-adaptive dual-source waste heat mode switching control method according to claim 5, characterized in that: If the control system determines within the rolling time window that none of the preset waste heat heating modes have been recorded as feasible states in the mode feasible domain throughout the entire time slice, it will maintain the current heating mode unchanged. The valve opening and circulating pump operating level corresponding to the current heating mode are limited to a preset safe state, and alarm information indicating that the mode feasible field is empty is output through the human-machine interface and the upper-level scheduling system.

7. The load-adaptive dual-source waste heat mode switching control method according to claim 1, characterized in that, S5 include: Within the rolling time window, the heating mode decision module selects preset waste heat heating modes for each time slot based on the mode feasible region, and generates a sequence of candidate heating modes in the order of the time slots. By calling the load heat demand trajectory and waste heat capacity fingerprint according to the time slice, the heat demand deviation, waste heat utilization ratio and mode switching cost of the candidate heating mode sequence in each time slice are calculated. Candidate heating mode sequences that do not meet the preset heat demand threshold and preset waste heat utilization threshold are eliminated. The remaining candidate heating mode sequences are calculated with a comprehensive evaluation score according to the preset evaluation weight. The candidate heating mode sequence with the highest comprehensive evaluation score is determined as the target heating mode in the current time slice, corresponding to the preset waste heat heating mode.

8. The load-adaptive dual-source waste heat mode switching control method according to claim 7, characterized in that: After determining the target heating mode, the heating mode decision module sends the target heating mode and its corresponding heating mode sequence identifier to the execution control module via the fieldbus. The execution control module verifies valve action commands and circulating pump action commands based on equipment operation constraints and real-time status. When it detects that the target heating mode does not meet the mode feasible domain, the constraint version identifier changes, or the action command exceeds the safe range, it returns feedback information containing error codes and execution version identifiers to the heating mode decision module. The heating mode decision module reselects the heating mode sequence based on the feedback information, and synchronizes the heating mode sequence and version identifier when it detects that the execution version identifier and the decision version identifier are inconsistent.

9. The load-adaptive dual-source waste heat mode switching control method according to claim 1, characterized in that, S6 include: When the target heating mode is selected, the control system establishes a mode switching session with a session identifier, and generates valve execution steps and circulation pump steps with action sequence numbers based on the heating mode sequence. The execution control module executes each step according to the action sequence number. After each action is completed, it judges whether the operation deviation exceeds the limit based on the key measurement points and equipment operation constraints. If the deviation exceeds the limit, a safety backoff strategy is executed. After the mode switching session is completed, the heating mode decision module adjusts the waste heat capacity fingerprint parameters based on the deviation between the actual heat exchange and the predicted waste heat capacity fingerprint value.

Images