A new energy photothermal cooperative optimization operation system based on regulation resource constraint identification

Abstract

The application discloses a new energy photothermal collaborative optimization operation system based on regulation resource constraint identification, and belongs to the technical field of intelligent dispatching of new energy power systems. The system comprises a data acquisition layer, a regulation resource constraint identification engine, an optimization decision layer, a collaborative regulation and control execution layer and a dynamic feedback loop. The regulation resource constraint identification engine establishes a dynamic model and maps physical constraints to a common time section according to various regulation resource operation state data collected by the data acquisition layer, so as to form a constraint boundary of a joint decision variable space. The optimization decision layer determines a target output sequence within the boundary and issues the target output sequence for execution. The dynamic feedback loop compares a deviation between an actual output value and a target output value and a measured response slope with a model response slope, and distinguishes between a transient execution deviation and a sustained deviation caused by a decrease in adjustable capacity, so as to finally realize unified expression of various regulation resource constraints and accurate identification of deviation causes, and improve dispatching executability and resource utilization.

Description

Technical Field

[0001] This invention relates to the field of intelligent dispatching technology for new energy power systems, and in particular to a new energy solar-thermal synergistic optimization operation system based on the identification of adjustment resource constraints. Background Technology

[0002] As the proportion of new energy sources such as wind and solar power in the power system continues to increase, the volatility and uncertainty of new energy output place higher demands on the system's power balance and regulation capabilities. The solar thermal power generation system's collector field, thermal storage unit, and thermoelectric conversion unit have the ability to transfer energy across time periods. The electric heating unit can realize the conversion of electrical energy into thermal energy, and the electric energy storage unit can quickly perform charge and discharge regulation. Coordinating and regulating the above resources is conducive to smoothing out the fluctuations of new energy and improving the utilization rate of renewable energy.

[0003] Different adjustable resources have different power limits, energy capacities, power change rates, response delays, and energy conversion relationships. Existing collaborative optimization methods typically generate scheduling plans under pre-defined equipment parameters and constraints, and update them on a rolling basis according to a fixed control cycle. However, due to continuous changes in resource operating temperature, energy storage status, equipment aging, and control link status, the adjustable capacity set in the model may deviate from the actual adjustable capacity of the execution unit. Furthermore, none of the aforementioned existing methods address a mechanism for uniformly expressing and dynamically updating the constraints of various adjustable resources. In actual operation, when the execution unit receives the adjustment command, its actual output may fail to reach the target in a timely manner. When outputting power, the execution deviation may be caused by the execution unit not completing its normal response process or by the decrease in the adjustable capability of the execution unit. If the adjustable boundary of the resource is directly reduced based solely on the execution deviation, the normal response delay may be misjudged as a decrease in capability, causing the available adjustable resources to be excluded too conservatively. If the execution deviation is compensated solely by issuing instructions in advance or repeatedly, the target output will still be generated beyond the actual capability when the execution unit has indeed experienced a decrease in capability, causing the deviation to accumulate continuously and potentially affecting the feasibility of resources with energy storage status in subsequent time segments. Existing methods lack the technical means to distinguish between the two causes of the above deviation and take corrective measures accordingly.

[0004] To address this, the present invention proposes a new energy solar-thermal synergistic optimization operation system based on the identification of resource constraints. Summary of the Invention

[0005] The purpose of this invention is to provide a new energy solar-thermal collaborative optimization operation system based on the identification of adjustable resource constraints, which solves the problems in the background technology of the difficulty in uniformly expressing various adjustable resource constraints, the difficulty in distinguishing the causes of execution deviations, and the poor executability of collaborative scheduling in the current new energy solar-thermal collaborative scheduling.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a new energy solar-thermal synergistic optimization operation system based on the identification of regulatory resource constraints, comprising: a data acquisition layer, used for: acquiring operational status data of each regulatory resource; a regulatory resource constraint identification engine, used for: establishing a dynamic model of each regulatory resource based on the operational status data, mapping the physical operational constraints of each regulatory resource to a common time section, and determining the constraint boundary of the joint decision variable space; an optimization decision layer, used for: determining the target output sequence of each regulatory resource based on the constraint boundary; and a collaborative control execution layer, used for: converting the target output sequence into resource control instructions and issuing them to... The corresponding execution unit; dynamic feedback loop, used to: determine the execution deviation and measured response slope based on the actual output value of each execution unit; when the execution deviation continuously exceeds the tolerance threshold and the absolute value of the measured response slope is lower than the absolute value of the model response slope of the corresponding dynamic model, it is determined that the adjustable capability of the execution unit has decreased, the adjustable power boundary of the subsequent time section of the corresponding execution unit is contracted and the optimization decision layer is triggered to update the target output sequence; when the execution deviation exceeds the tolerance threshold and the absolute value of the measured response slope is not lower than the absolute value of the model response slope, it is determined to be a normal response process, the adjustable power boundary is maintained and the issuance time of the corresponding resource control command is adjusted according to the execution deviation.

[0007] Furthermore, each regulating resource includes at least two of the following: a heat collection field, a heat storage unit, a thermoelectric conversion unit, an electric heating unit, and an electric energy storage unit. The operating status data includes data characterizing the current output, energy storage status, adjustable power range, power change rate, and response delay of each regulating resource.

[0008] Furthermore, the regulating resource constraint identification engine normalizes the power and energy variables according to the rated capacity of each regulating resource, and maps the corresponding executable output range to a common time section according to the response delay of each regulating resource, so as to obtain the standardized power adjustable range and standardized energy accumulation relationship of each regulating resource under the common time section.

[0009] Furthermore, the regulatory resource constraint identification engine extracts the upper and lower limits of state variables, the upper and lower limits of control variables, the rate of change limits of adjacent time sections, and the energy or power balance relationship from the dynamic model of each regulatory resource. The extracted physical operation constraints are configured into the joint decision variable space according to the regulatory resources and time sections to form a constraint set representing the temporal coupling relationship of each regulatory resource, and the constraint boundary is determined according to the constraint set.

[0010] Furthermore, the dynamic feedback loop determines the measured response slope based on the actual output change of the execution unit at adjacent sampling times and the sampling time interval, and determines the model response slope based on the expected output change calculated by the corresponding dynamic model within the same time interval according to its own response characteristics. The model response slope reflects the actual output change rate that the device can achieve under a given control input.

[0011] Furthermore, when the duration of the execution deviation exceeding the tolerance threshold reaches the response delay of the corresponding execution unit, the dynamic feedback loop compares the absolute value of the measured response slope with the absolute value of the model response slope to distinguish between transient execution deviation caused by the incomplete response process and continuous execution deviation caused by the decrease in the adjustability of the execution unit.

[0012] Furthermore, when the absolute value of the measured response slope is lower than the absolute value of the model response slope, the dynamic feedback loop determines the boundary contraction amount based on the difference between the absolute value of the measured response slope and the absolute value of the model response slope, and applies the boundary contraction amount to the upper or lower power boundary of the corresponding execution unit according to the direction of the execution deviation.

[0013] Furthermore, when the corresponding execution unit is an adjustable resource with an energy storage state, the adjustable resource constraint identification engine updates the energy boundary of subsequent time sections along the energy state recursion relationship based on the contracted adjustable power boundary and the measured energy storage state of the corresponding execution unit, and redetermines the constraint boundary of the joint decision variable space based on the updated power boundary and energy boundary.

[0014] Furthermore, when the execution deviation exceeds the tolerance threshold and the absolute value of the measured response slope is not lower than the absolute value of the model response slope, the dynamic feedback loop will identify the execution deviation as a transient execution deviation, and based on the estimated remaining response time required for the actual output to reach the target output, will advance the issuance time of the corresponding resource control command in the next control cycle, while maintaining the adjustable power boundary and energy boundary of the corresponding execution unit.

[0015] Furthermore, after the constraint boundary is updated, the optimization decision layer determines the adjustment amount that the corresponding execution unit failed to achieve, selects candidate execution units from other adjustment resources whose response delay does not exceed the remaining time of the current control cycle and are located within the updated constraint boundary, and allocates the failed adjustment amount according to the remaining adjustment margin of each candidate execution unit.

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention proposes a new energy solar-thermal synergistic optimization operation system based on the identification of regulatory resource constraints. In existing technologies, the synergistic scheduling of new energy solar-thermal resources suffers from problems such as difficulty in uniformly expressing various regulatory resource constraints, difficulty in distinguishing the causes of execution deviations, and poor executability of synergistic scheduling. This invention, however, establishes a dynamic model of each regulatory resource through a data acquisition layer, a regulatory resource constraint identification engine, an optimization decision layer, a synergistic control execution layer, and a dynamic feedback loop. It maps power, energy, rate of change, and response delay constraints to a common time section to form a joint decision variable space. This is achieved by addressing the issue of continuous exceeding of execution deviations for extended periods, thereby optimizing the operation of the corresponding resource. After the response delay, the causes of the deviation are identified by comparing the measured response slope with the model response slope. Different correction paths are executed for different causes of deviation. For resources with reduced capacity, the power boundary in the corresponding adjustment direction is contracted in a targeted manner. For resources with no reduction in response capacity, the original boundary is retained and the command timing is corrected. The power boundary change of energy storage resources is transmitted to subsequent time sections along the energy state relationship. The unrealized adjustment amount is transferred within the updated constraint boundary to reduce the accumulation of local execution deviations into subsequent control cycles. Finally, unified modeling of various adjustment resources and accurate identification of deviation causes are achieved, improving the executability of collaborative scheduling results and resource utilization, and reducing the false contraction of resources with normal response. Attached Figure Description

[0017] Figure 1 A schematic diagram of the overall architecture of the new energy solar-thermal synergistic optimization operation system based on resource constraint identification provided by the present invention; Figure 2 This is a schematic diagram illustrating the process by which the resource constraint recognition engine provided by the present invention maps various resource constraints to a common time section. Figure 3 This is a schematic diagram of the dynamic feedback loop deviation cause identification and processing logic provided by the present invention; Figure 4 This is a schematic diagram illustrating the energy boundary update and adjustment transfer of energy storage resources after adjustable power boundary contraction, as provided by 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] To address the technical challenges of improving the feasibility and resource utilization of coordinated solar and thermal energy scheduling, such as... Figure 1-4As shown, the following preferred technical solution is provided: a new energy solar-thermal synergistic optimization operation system based on the identification of regulatory resource constraints, comprising: a data acquisition layer, used to: collect the operating status data of each regulatory resource. For the solar collector field, the operating status data collected by the data acquisition layer includes the heat absorption medium flow rate, inlet temperature, outlet temperature, and current heat collection power; for the thermal storage unit, the operating status data collected by the data acquisition layer includes the thermal storage medium temperature, heat storage capacity, charging power, and releasing power; for the thermoelectric conversion unit, the operating status data collected by the data acquisition layer includes steam parameters, unit electrical power, allowable load increase / decrease rate, and start / stop status; for the electric heating unit, the operating status data collected by the data acquisition layer includes input electrical power, output thermal power, and allowable adjustment range; for the electric energy storage unit, the operating status data collected by the data acquisition layer includes charging / discharging power, state of charge, allowable charging / discharging power, and... Operating mode; The data acquisition layer also performs time alignment on the timestamped operating status data. For operating status data that is missing or exceeds the device's range, the data acquisition layer does not use the operating status data to update the dynamic model of the corresponding adjustment resource. Instead, it retains the operating status data of the corresponding adjustment resource that has most recently passed the validity check and outputs the valid operating status data of the corresponding adjustment resource to the adjustment resource constraint recognition engine. After the operating status data of the corresponding adjustment resource is restored to validity, the data acquisition layer uses the continuously obtained valid operating status data to update the dynamic model state of the corresponding adjustment resource to avoid abrupt changes in the constraint boundary of the corresponding adjustment resource caused by a single abnormal sampling point.

[0020] The resource constraint identification engine is used to: establish dynamic models of each regulatory resource based on the operating status data, map the physical operating constraints of each regulatory resource to a common time segment, and determine the constraint boundaries of the joint decision variable space; specifically, the resource constraint identification engine forms the power range and energy range that each regulatory resource can execute in the prediction time domain, the dynamic model of each regulatory resource describes the relationship between the state of the corresponding regulatory resource and the control input and time, and the dynamic model of each regulatory resource adopts a state recursive model, a piecewise linear model, a characteristic curve model, or a response model identified from the equipment trial operation data. The specific form of the dynamic model of each regulatory resource is selected according to the type of the corresponding regulatory resource. The resource constraint identification engine requires that the dynamic model of each regulatory resource can output the executable output range, state boundary, and expected response process of the current time segment and subsequent time segments.

[0021] For different types of regulating resources, their state variables, control variables, and constraint relationships each have their own characteristics. The state variables of the heat collection field include the temperature of the heat absorption medium, the flow rate, and the available heat collection power. The control variables are the flow rate setting and the mirror field focusing ratio. The constraint relationships include temperature range, flow rate range, and heat collection power change rate constraints. The state variables of the heat storage unit include the heat storage capacity, the hot and cold end temperatures, and the heat charging and discharging status. The control variables are the heat charging power and the heat discharging power. The constraint relationships include upper and lower limits of heat storage capacity, heat charging and discharging power limits, and heat loss constraints. The state variables of the thermoelectric conversion unit include unit output, steam status, and operating status constraints. The control variables are the target electric power and load increase / decrease commands. The constraint relationships include minimum stable output, maximum output, ramp-up limit, and start / stop constraints. The state variables of the electric heating unit include the current input power and available status. The control variable is the target input power. The constraint relationships include upper and lower limits of power, change rate, and conversion capacity constraints. The state variables of the electric energy storage unit include the state of charge, charge / discharge power, and available status. The control variable is the target charge / discharge power. The constraint relationships include the state of charge range, power range, energy recursion relationship, and mode mutual exclusion constraints.

[0022] To enable regulation resources with different dimensions to enter the same joint decision variable space, the regulation resource constraint identification engine normalizes power and energy variables according to the rated capacity or current available capacity of each regulation resource. For regulation resources with response delays less than one time segment length, the regulation resource constraint identification engine configures the executable power range of the corresponding regulation resource into the joint decision variable space starting from the current time segment or the next time segment. For regulation resources with response delays spanning one or more time segment lengths, the regulation resource constraint identification engine marks the interval between the instruction issuance time and the expected start time of response as an unaddressable regulation interval and marks the executable power range as an unaddressable regulation interval. The force interval is extended to the time segment where the corresponding regulating resource can actually respond; the regulating resource constraint identification engine establishes the corresponding variable positions in the joint decision variable space according to each regulating resource and each time segment, and configures the upper and lower limits of state variables, the upper and lower limits of control variables, the rate of change limit of adjacent segments, the energy balance relationship and the transformation relationship to the corresponding variable positions. For thermal storage units and electric energy storage units, the regulating resource constraint identification engine also establishes the recursive relationship between the energy state of the current time segment and the energy state of subsequent time segments. The constraint relationships of all regulating resources together form the constraint set of the joint decision variable space, and the variable range that satisfies the constraint set constitutes the constraint boundary of the joint decision variable space.

[0023] The optimization decision layer is used to: determine the target output sequence of each regulating resource based on the constraint boundary; the optimization decision layer receives data on the predicted output of new energy sources, load demand, or grid regulation demand, and determines the target output sequence of each regulating resource within the constraint boundary of the joint decision variable space. The optimization objectives of the optimization decision layer include reducing the amount of new energy curtailment, reducing the tracking deviation of the plan, reducing energy conversion losses, and maintaining the sustainability of energy storage status. Different operational objectives can be prioritized according to the station operation tasks, but any target output value in the target output sequence must be within the constraint boundary output by the regulating resource constraint identification engine.

[0024] The coordinated control execution layer is used to: convert the target output sequence into resource control instructions and issue them to the corresponding execution units; the coordinated control execution layer reads the target output value corresponding to the current control cycle in the target output sequence, converts the target output value into an active power or thermal power setpoint for execution units that can be directly mobilized, determines the instruction preparation time based on the response delay for execution units that need to be started in advance, and performs power return to zero and mode permission checks before switching between charging / discharging or charging / discharging states for execution units with mutually exclusive modes; each resource control instruction includes at least a resource identifier, target output value, planned effective time, and corresponding control cycle information, and each execution unit returns the instruction reception status and actual output value to the coordinated control execution layer. The dynamic feedback loop establishes a correspondence between the target output value and the actual output value based on the planned effective time to avoid incorrectly using the target output value that has not yet taken effect for execution deviation calculation.

[0025] The dynamic feedback loop is used to: determine the execution deviation and measured response slope based on the actual output value of each execution unit; when the execution deviation continuously exceeds the tolerance threshold and the absolute value of the measured response slope is lower than the absolute value of the model response slope of the corresponding dynamic model, the dynamic feedback loop shrinks the adjustable power boundary of the corresponding execution unit in subsequent time sections and triggers the optimization decision layer to update the target output sequence; when the execution deviation exceeds the tolerance threshold and the absolute value of the measured response slope is not lower than the absolute value of the model response slope, the adjustable power boundary is maintained and the issuance time of the corresponding resource control command is adjusted according to the execution deviation.

[0026] The dynamic feedback loop calculates the difference between the target output value and the actual output value at each sampling moment, and uses this difference as the execution deviation. The direction of the execution deviation indicates whether the actual output value of the execution unit is lower or higher than the target output value. The tolerance threshold is determined based on the measurement accuracy of the adjustment resources, the normal fluctuation range, and the minimum adjustable step size. To avoid frequent switching caused by noise, the dynamic feedback loop only enters the deviation cause identification process when the execution deviation exceeds the tolerance threshold during continuous sampling. The dynamic feedback loop calculates the difference between the actual output change of the corresponding execution unit at adjacent sampling moments and the sampling time interval. The measured response slope is obtained; based on the expected output change of the corresponding execution unit's dynamic model within the same time interval, considering its inherent ramp rate, response delay, and energy transfer inertia, the model response slope is determined. The model response slope represents the maximum or typical response rate that the equipment can actually achieve under the current operating conditions, and is independent of the target output change required in the control command (which may be a step command), thus avoiding mistaking the steep change rate of the command as the normal response rate of the equipment; the average or median value of the measured response slope values ​​at multiple consecutive sampling times is taken as the measured response slope for comparison.

[0027] The dynamic feedback loop does not immediately shrink the adjustable power boundary based on the slope difference after the command is issued. First, it counts the duration for which the execution deviation continuously exceeds the tolerance threshold. When the duration has not yet reached the response delay of the execution unit, the execution deviation is kept in the observation state. Only when the duration reaches the response delay of the execution unit is the absolute value of the measured response slope compared with the absolute value of the model response slope. Through the setting of the above comparison timing, the dynamic feedback loop separates the command propagation stage, the equipment preparation stage, the normal response waiting stage, and the adjustable capability judgment stage.

[0028] For deviation cause identification and subsequent processing, the dynamic feedback loop executes different branches according to the following judgment conditions: If the execution deviation does not continuously exceed the tolerance threshold, it is determined to be a normal tracking state, and the original adjustable power boundary and command plan remain unchanged; if the execution deviation exceeds the tolerance threshold but its duration has not yet reached the response delay of the execution unit, the current state is identified as a transient process to be observed, no corrective measures are taken, and sampling and monitoring continue; when the execution deviation duration has reached the response delay of the execution unit, the measured response slope is further compared with the model response slope. If the absolute value of the measured response slope is lower... If the absolute value of the model response slope indicates a decrease in the adjustability of the execution unit, it is determined to be a continuous execution deviation. In this case, the power boundary of the corresponding adjustment direction is directionally contracted based on the difference in the absolute value of the slope, and the optimization decision layer is triggered to re-optimize and transfer the unrealized adjustment amount. If the absolute value of the measured response slope is not lower than the absolute value of the model response slope, it is determined to be a transient execution deviation. That is, the adjustability of the corresponding execution unit has not decreased, but there is a lag in the response timing, which is a normal response process. In this case, the adjustable power boundary remains unchanged, and only the issuance time of the corresponding resource regulation command in the next control cycle is adjusted to compensate for the transient execution deviation.

[0029] When the absolute value of the measured response slope is lower than the absolute value of the model response slope, the dynamic feedback loop determines the boundary contraction amount based on the absolute difference between the measured response slope and the model response slope, and applies the boundary contraction amount to the upper or lower power boundary of the corresponding execution unit according to the direction of the execution deviation. If the target output value requires an upward adjustment but the actual output value is lower than the target output value, the dynamic feedback loop applies the boundary contraction amount to the upper power boundary of the corresponding execution unit. If the target output value requires a downward adjustment but the actual output value is higher than the target output value, the dynamic feedback loop applies the boundary contraction amount to the lower power boundary of the corresponding execution unit. The adjustable power boundary in the direction unaffected by the execution deviation remains unchanged.

[0030] For regulated resources with energy storage states, including thermal storage units and electrical energy storage units, after the regulated resource constraint identification engine shrinks the adjustable power boundary in the dynamic feedback loop, it updates the achievable energy range for each subsequent time segment according to the energy input-output relationship, conversion efficiency relationship, and self-loss relationship, based on the shrunken adjustable power boundary and the current measured energy storage state of the corresponding execution unit. The regulated resource constraint identification engine then redetermines the constraint boundary of the joint decision variable space based on the updated adjustable power boundary and updated energy boundary. When the execution deviation exceeds the tolerance threshold and the absolute value of the measured response slope is not lower than the absolute value of the model response slope, the dynamic feedback loop identifies the execution deviation as a transient execution deviation. Based on the current actual output value, target output value, and measured response slope, the dynamic feedback loop estimates the remaining response time required for the corresponding execution unit to reach the target output value, and adjusts the issuance time of the corresponding resource regulation command in the next control cycle forward based on the remaining response time. The advance of the issuance time does not exceed the maximum allowable advance for the corresponding execution unit. The start time is set ahead of schedule and no earlier than the time when the current control cycle allows modification of the instruction. The dynamic feedback loop maintains the adjustable power boundary and energy boundary of the corresponding execution unit unchanged while adjusting the instruction issuance time. When the corresponding execution unit cannot achieve the original target output value due to the contraction of the adjustable power boundary, the optimization decision layer calculates the adjustment amount that the corresponding execution unit failed to achieve. The optimization decision layer selects candidate execution units from other adjustment resources. The candidate execution units must simultaneously meet the following conditions: the response delay does not exceed the remaining time of the current control cycle, the proposed adjustment amount is within the updated constraint boundary, and the increase in adjustment amount will not cause the subsequent energy state to exceed the boundary. The optimization decision layer allocates the failed adjustment amount proportionally or according to priority based on the remaining adjustment margin of each candidate execution unit. After each allocation, the optimization decision layer re-verifies the adjustable power boundary, rate of change limit, and energy boundary of each candidate execution unit. If the sum of the available adjustment margins of all candidate execution units is still insufficient to cover the failed adjustment amount, the optimization decision layer outputs the remaining gap.

[0031] Before being put into operation, the dynamic model of each regulating resource is established based on the rated parameters of the equipment, the set values ​​of the controller, and the trial operation records. During operation, the dynamic model of each regulating resource is calibrated using the measured state data that has been verified for effectiveness. The dynamic model calibration is used to correct the initial state and the executable range without changing the hard constraints given by the equipment protection system. Any adjustable power boundary obtained through dynamic feedback loop updates must not exceed the safety upper limit specified by the equipment manufacturer, the protection limit specified in the operating procedure, or the interlocking conditions of the field controller.

[0032] The dynamic feedback loop also sets the recovery conditions for the adjustable power boundary. When the execution deviation of the corresponding execution unit remains within the tolerance threshold in multiple subsequent control cycles, and the measured response slope recovers to the allowable range of the model response slope of the corresponding dynamic model, the resource constraint identification engine expands the adjustable power boundary of the corresponding execution unit in the corresponding direction in a step-by-step recovery manner. The recovered adjustable power boundary must still not exceed the hard constraint of the device.

[0033] Regarding parameter settings, the control cycle length, sampling interval length, tolerance threshold, and allowable slope difference are set separately according to the type of regulating resource. The electric energy storage unit uses the first control cycle length, while the thermal storage unit and thermoelectric conversion unit use the second control cycle length. The first control cycle length is shorter than the second control cycle length. The common time section uses the basic time granularity, representing the response delay of the slow-response regulating resource through multiple basic time sections. The tolerance threshold is greater than the comprehensive error of the measurement system and the normal steady-state fluctuation range of the corresponding execution unit. The response delay is the response delay value in the equipment's technical parameters, or an updated response delay value based on the statistical results between the historical command issuance time and the actual output starting to stabilize. The response delay value is used for the next control cycle and does not retrospectively change the already completed deviation judgment. The allowable range of the model response slope is based on the current dynamic model output model response slope, and is expanded to form a judgment interval after considering uncertain factors such as measurement noise and operating condition fluctuations. It is not a third physical quantity independent of the model response slope. The allowable range of the model response slope is formed based on the recent normal operation records of the corresponding adjustment resources and is limited by the rated change rate of the equipment. When the measured response slope is within the allowable range, the dynamic feedback loop determines that the absolute value of the measured response slope is not lower than the absolute value of the model response slope. When the measured response slope is lower than the lower limit of the allowable range, the dynamic feedback loop determines that the absolute value of the measured response slope is lower than the absolute value of the model response slope.

[0034] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A new energy solar-thermal synergistic optimization operation system based on resource constraint identification, characterized in that, include: The data acquisition layer is used to collect operational status data of various regulatory resources. The adjustment resource constraint identification engine is used to: establish dynamic models of each adjustment resource based on the operating status data, map the physical operating constraints of each adjustment resource to a common time segment, and determine the constraint boundary of the joint decision variable space; the optimization decision layer is used to: determine the target output sequence of each adjustment resource based on the constraint boundary; the collaborative control execution layer is used to: convert the target output sequence into resource control instructions and issue them to the corresponding execution units; the dynamic feedback loop is used to: determine the execution deviation and the measured response slope based on the actual output value of each execution unit. When the execution deviation continuously exceeds the tolerance threshold and the absolute value of the measured response slope is lower than the absolute value of the model response slope of the corresponding dynamic model, it is determined that the adjustable capability of the execution unit has decreased, the adjustable power boundary of the subsequent time segment of the corresponding execution unit is contracted, and the optimization decision layer is triggered to update the target output sequence. When the execution deviation exceeds the tolerance threshold and the absolute value of the measured response slope is not lower than the absolute value of the model response slope, it is determined to be a normal response process, the adjustable power boundary is maintained, and the issuance time of the corresponding resource control instructions is adjusted according to the execution deviation.

2. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 1, characterized in that, The regulating resources include at least two of the following: a thermal collector field, a thermal storage unit, a thermoelectric conversion unit, an electric heating unit, and an electric energy storage unit. The operating status data includes data characterizing the current output, energy storage status, adjustable power range, power change rate, and response delay of each regulating resource.

3. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 1, characterized in that, The regulating resource constraint identification engine normalizes the power and energy variables according to the rated capacity of each regulating resource, and maps the corresponding executable output range to a common time section according to the response delay of each regulating resource, so as to obtain the standardized power adjustable range and standardized energy accumulation relationship of each regulating resource under the common time section.

4. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 3, characterized in that, The regulatory resource constraint identification engine extracts the upper and lower limits of state variables, the upper and lower limits of control variables, the rate of change limits of adjacent time sections, and the energy or power balance relationship from the dynamic model of each regulatory resource. The extracted physical operation constraints are configured into the joint decision variable space according to the regulatory resources and time sections to form a constraint set representing the temporal coupling relationship of each regulatory resource, and the constraint boundary is determined according to the constraint set.

5. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 1, characterized in that, The dynamic feedback loop determines the measured response slope based on the actual output change of the execution unit at adjacent sampling times and the sampling time interval, and determines the model response slope based on the expected output change calculated by the corresponding dynamic model within the same time interval according to its own response characteristics. The model response slope reflects the actual output change rate that the device can achieve under a given control input.

6. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 5, characterized in that, When the duration of the execution deviation exceeding the tolerance threshold reaches the response delay of the corresponding execution unit, the dynamic feedback loop compares the absolute value of the measured response slope with the absolute value of the model response slope to distinguish between transient execution deviation caused by the incomplete response process and continuous execution deviation caused by the decrease in the adjustability of the execution unit.

7. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 6, characterized in that, When the absolute value of the measured response slope is lower than the absolute value of the model response slope, the dynamic feedback loop determines the boundary contraction amount based on the difference between the absolute value of the measured response slope and the absolute value of the model response slope, and applies the boundary contraction amount to the upper or lower power boundary of the corresponding execution unit according to the direction of the execution deviation.

8. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 7, characterized in that, When the corresponding execution unit is an adjustable resource with an energy storage state, the adjustable resource constraint identification engine updates the energy boundary of subsequent time sections along the energy state recursion relationship based on the contracted adjustable power boundary and the measured energy storage state of the corresponding execution unit, and redetermines the constraint boundary of the joint decision variable space based on the updated power boundary and energy boundary.

9. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 6, characterized in that, When the execution deviation exceeds the tolerance threshold and the absolute value of the measured response slope is not lower than the absolute value of the model response slope, the dynamic feedback loop will identify the execution deviation as a transient execution deviation, and based on the estimated remaining response time required for the actual output to reach the target output, will advance the issuance time of the corresponding resource control command in the next control cycle, while maintaining the adjustable power boundary and energy boundary of the corresponding execution unit.

10. The new energy solar-thermal synergistic optimization operation system based on resource constraint identification according to claim 7 or 8, characterized in that, After the constraint boundary is updated, the optimization decision layer determines the adjustment amount that the corresponding execution unit failed to achieve, selects candidate execution units from other adjustment resources whose response delay does not exceed the remaining time of the current control cycle and are located within the updated constraint boundary, and allocates the failed adjustment amount according to the remaining adjustment margin of each candidate execution unit.

Images